combinepdf (5)

A STUDY ON BIOMETRIC IMAGE PROCESSING 

J. Arockia Jenifer & S. Lakshmi 

Fatima College(Autonomous), Madurai 

ABSTRACT 

Biometric image processing and recognition, which can be used in attendance, like 

employee attendance, student attendance. It is more secure for user access, e-commerce and other 

security applications. This includes the addition of the different options such as fingerprint based 

attendance system, face recognition attendance systems. 

This biometric scan can solve issues pertaining to information security. Fundamental 

challenges encountered by biometric systems in real world applications. Biometric technology 

used to measure and analyze personal characteristics, these characteristics include fingerprints, 

voice patterns, hand measurements. 

Fingerprints are considered to be the best and fastest method for biometric identification. 

The biometric identification systems are widely used for unique identification, mainly for 

verification and identification. Using fingerprint based attendance system has been introduced for 

automatically monitoring and calculating the student attendance in a class. 

INTRODUCTION 

Image processing is a method to perform some operations on an image, in order to get an 

enhanced image or to extract some useful information from it. It is a type of signal processing in 

which input is an image and output may be image characteristics associated with the image. 

Analyzing and manipulating the image. 

Digital Image Processing is the technology of manipulating the groups of bits to enhance 

the quality of the image or create different perspectives or to extract information from the image 

digitally, with the help of computer algorithms. Digital image is how computer record the 

pictures. The smallest unit is called pixel, normally consists of a 0-255 value for gray images. For 

color images, each pixel has three 0-255 values, representing RGB. 

Digital image processing visualization observes the objects, which are not visible. Image 

sharpening and restoration is used to create a better image. Measurement of pattern measures 

various objects in an image. Biometric image processing and recognition can be used in 

attendance. Biometric attendance systems are quickly gaining in most offices and institutions. 

It can reduce problems such as the presence of the missing paper and easily damaged. 

With this system can replace the existing manual system to a more systematic.

In fingerprint attendance system with access control, that not only allows you to login and 

out times of the employees, but also enables you prevent unauthorized entries into the workplace. 

This makes easier for both the employee and the business as work hours are logged automatically 

when the employee enters and leaves the office. This eliminates the possibility of timesheets 

getting lost or manipulated; it also saves a lot of time. In biometrics, image processing is required 

for identifying and individual whose biometric image is stored in the database previously, faces, 

fingerprints, etc., are image based biometrics, which require image processing and pattern 

recognition techniques. 

TECHNOLOGY USED 

1. Field used 

It can be used in institutions, colleges, companies and many developing fields. An 

automated system eliminates the need for paper tracking and instead makes use of touch screens, 

magnetic stripe cards. This makes easier for both the employees and students. Automated 

fingerprint verification is a closely related technique used in applications such as attendance and 

access control systems. It compares the basic fingerprint patterns between a previously stored 

template and a candidate fingerprint. 

2. Fingerprint Process 

Biometric technologies can provide a means for uniquely recognizing humans based upon 

one or more physical or behavioral characteristics and can be used to establish or verify personal 

identity of individuals previously enrolled. 

Fingerprints are considered to be the best method for biometric identification. They are 

secure to use, unique for every person and do not change in one’s life time. A fingerprint 

recognition system operates either in verification mode or in identification mode. Automated 

fingerprint identification is the process of automatically matching one or many unknown 

fingerprints against a database of known and unknown prints. 

Finger scanners consist of 

‣ A reader or scanning device. 

‣ Software that coverts the scanned information into digital form and compares 

matchpoints. 

‣ A database that stores the biometric data for comparison.

Binarization converts gray scale image into binary image by image by fixing the threshold 

value. The pixel values above and below the threshold are set to 1 and 0 respectively. 

It’s the most critical task in the fingerprint matching system. The binarized image is thinned using 

block filter to reduce the thickness of all ridge lines to a single pixel width to extract minutiae 

points effectively. Thinning preserves outermost pixels by placing white pixels at the boundary of 

the image, as a result first five and last five rows, first five and last five columns are assigned 

value of one. 

The minutiae location and the minutiae angles are derived after minutiae extraction. To 

compare the input fingerprint data with the template data minutiae matching is used. Its use a 

minute features on the finger. 

The major minutiae features are ridge ending, bifurcation, and short ridge. The ridge 

ending is the point at which a ridge terminates. Single ridge splits into two ridges. Short ridges 

are ridges which are significantly shorter than the average ridge length on the fingerprint. 

Minutiae and patterns are very important in the analysis of fingerprints since no two fingers have 

been shown to be identical. During the matching process, each input minutiae point is compared 

with template minutiae point. 

3. Finger Based Attendance System 

Student identification should be done by student’s finger print, for identification, the 

device scans the ridge and edge of the finger and creates a template. The system searches all the 

templates that are stored in the system database and matches with each saved template. Student’s 

fingerprints from the fingerprint scanner used as input. The fingerprint scanner can read 

fingerprints of any or more fingers of the both hands. The basic information was stored in the 

particular table and the fingerprints were in the template data table.

CONCLUSION 

This paper presented the related works and performance analysis for fingerprint biometric. 

This paper mainly comprised of development of attendance management system and fingerprint 

identification system. Attendance system management was made automated and online. There are 

some limitations of the fingerprint technology. In identification mode, biometric sample is taken 

for future recognition purpose and in verification mode. It is platform independent and results are 

very accurate. It gives a real time security system along with portability and low cost. 

REFERENCES 

1. Zhang Yongqiang and Liu Ji, The design of wireless fingerprint attendance system, 

International Conference on CommunicationTechnogy, 2010. 

2. Younhee Gil, Access Control System with high level security using fingerprints,IEEE 

the 32nd Applied Imagery Pattern Recognition Workshop (AIPR 09) 

3. Jain, A.K., Hong, L., and Bolle, R(1997), On-Line Fingerprint Verification, IEEE 

Trans. On Pattern Anal And Machine Intell.

MOBILE SENSING ON INTERNET OF THINGS 

R.Astalakshmi & M.H.Saied Sharmila 

MAM College, Trichy 

ABSTRACT 

The Internet of Things (IOT) is rising rapidly and its interconnection with mobile devices 

or Web is increasing. Mobile devices connected with sensors provide more developed services, 

enhance the user knowledge, experience, awareness and encourage better living. This paper is a 

survey paper which provides various applications or field of mobile sensing when combined with 

IOT or Web technology. The study has been done over 60-70 papers, which comprised of most 

relevant information in the domain, categorize these papers into eight different categories based 

on their area of application (health, transportation, games & sports, and agriculture), the nature of 

interface (crowd sensing, feedback, control). The challenges and problems have been analyzed 

and discussed. Finally, suggestions were provided to enhance the use of mobile sensing in various 

fields. Keywords—Internet of Things (IOT), mobile sensing, Radio Frequency Identification 

(RFID), QR sensing, barcodes, NFC, Web of Things (WOT). 

INTRODUCTION 

The present period of remote system sensors, radio recurrence identification(RFID) and 

short-run remote correspondence have allowed the Internet to get to or to experience an installed 

processing. Ordinary objects of our encompassing are joined to sensors, getting to be 

interconnected and are remarkably tended to, permits association among them through Internet 

made the Internet of Things a reality. Porting in IPv6 (128 bits) gives the extensive tending to go 

which gives the office of joining non-living condition with advanced world (Mobiles, web 

applications and so forth.), causes the IOT to develop at quicker rate. This developing innovation 

is upheld by the Web of Things (WOT) which additionally characterizes the Web systems to 

interface this age with physical items.The IOT interconnect the distinctive gadgets at the Network 

layer while the WOT is stretching out the Web practice to port the gadgets into it. The connecting 

of portable registering with IOT and WOT has expanded with the quick ascent of cell phones 

with multi-sensors. This gives the higher applications to estimating temperature, ascertaining 

increasing speed and checking areas or record recordings or catch photographs.There are a few 

study papers identified with IOT and Mobile detecting, however we have reviewed the quick 

developing interconnection between portable detecting and Internet of Things or Web. We have 

experienced such a significant number of works in the field of versatile detecting and Internet of

Things. Lastly determined different classes that clarified the work or versatile applications in 

particular field. These applications have experienced with numerous issues and difficulties, which 

have been examined in last segment. 

CLASSIFICATION OF APPLICATION 

The mobile phone has become a mandatory aspect of our life. The mobile device has 

extended its use in areas for text messages, taking pictures and even detecting our location (GPS 

built in). The internet of things and the web of things (IOT, WOT) have made it possible for the 

user to get interacted with the physical world through mobile phone and could sense everything 

with the sensing application of the mobile phone. The work relates to mobile technologies can 

also be classified in areas like: 

1) Crowd sensing 

The term crowd sensing is a method wherein the data is collected and shared with the help 

of mobile devices. Due to the ubiquitous presence of mobile devices, it has become an enticing 

method for the businesses and companies including Google, Face book, etc. for providing 

services. 

2) Control 

Controlling the appliances is again an area where the sensing factor has played an 

important role. The home appliances are easily controlled by the sensors how far we are from 

home. There are other areas too like offices and factories where it can be a daily need. 

3) Health 

The wellbeing related portable application is frequently utilized strategy generally for the 

seniority ShivamJuyalNavyata Joshi Rahul Chauhan Dept. of Computer Science and Dept. of 

Electronics and Communication Dept. of Electronics and Communication Engineering, Graphic 

Era Hill University Engineering, Graphic Era Hill University Engineering, Graphic Era Hill 

University Dehradun, India Dehradun, India Dehradun, India shivamjuyal19@gmail.com 

navyata12joshi@gmail.com chauhan14853@gmail.com individuals today. The sensors identify 

the body movement and transfer it to the web for breaking down it and after that give the 

criticism.

4) Feedback applications 

These incorporate versatile applications that offer criticism to the client about their 

ecological conditions. These applications can decrease the undeniably negative impacts which 

can be because of their exorbitant ecological utilization. 

5) Agriculture 

To improve the productivity and get intact with the customer’s mobile applications have 

been a boom for the farmers. It even helps to check the productivity and manage the livestock. 

This is known as Smart Farming. 

6) Games and Sports 

The different physical sensors used during sports activity gives the information about the 

athlete performance, checking the heart rate measure the speed and the separation secured. The 

versatile moreover shares web diversions which help them with rivaling the buddies enduring the 

troubles 

7) Interaction with Surroundings 

There can be seen innumerable applications today which have related the world. The long 

range relational correspondence goals share the information with the all inclusive community and 

swell our social messages. 

8) Transportation 

The identifying features of the mobile phone are furthermore used to take a gander at it the 

fitting ceasing domains and all through help the action delays, road condition and caution the 

setbacks. In this stage, each application is given and discussed most weighty works that have been 

done in the individual field. 

‣ Versatile SENSING 

‣ Transportation Control 

‣ Recreations and Sports 

‣ Wellbeing

‣ Group detecting 

‣ Criticism 

‣ Horticulture 

A. Group detecting 

Gathering identifying is on a very basic level an amassing of data and decoding it. It 

basically weights on identifying the zone where they live. It is a participation of the social 

occasion of people, sharing their data which help for facing the environmental challenges. With 

the use of recognizing PDAs, it will help us with investigating the parts of our lives. It's a scene 

of get-together and sharing the area information and separate it. In a couple of utilizations like 

Noise Tube, mobile phones work as a fuss sensor to evaluate the uproar presentation of a man in 

the customary condition. The versatile identifying is fundamental for the customers to get 

possessed with different activities and it in like manner helps in getting aware of their condition 

and find the solutions for its headway. 

B. Control 

The mobile phones can in like manner be used as a source to control diverse physical 

devices from wherever at whatever point. A champion among the most surely understood 

conventional outlines is the home computerization which may join controlling the lights, 

ventilation, circulating air through and cooling and furthermore the security structure. This 

methodology can in like manner be used as a piece of different distinctive fields. The Internet of 

things has given the resolute quality to identify and screen the things inside a framework. The 

advances of the web have been giving moved other options to the home automation. The 

contraptions can be identified with the help of Wi-Fi and subsequently through the web the 

devices can be controlled. A flexible application system used to control and adjust the lights and 

aficionados of the house is presented in. This is a measure to save the essentialness and use 

contraptions more supportively. There are various more game plans made to work for the home 

and the building computerization through the employments of web and web. An Ocean is a 

computerization advantage that controls the earth of the house through the mobile phone. 

Moreover, even the advancement of plants can be distinguished with the help of sensors and 

mobile phone to manage parameters like stickiness, temperature, and watering office. Therefore, 

with the help of convenient applications, the physical contraptions can be controlled satisfactorily 

giving simplicity and moreover the individual fulfillment. Essentialness saving can in like manner

e said as a factor which is seen here for our advantage and pleasant life in the present and 

moreover later on. 

C. Wellbeing 

In a specific order, flexible applications related the body sensors and the web or telephone 

sensors, and are utilized to perceive some issue by assessing some strange lead of the body part. 

Body sensors check the flourishing status and send the standard to the telephones, which is shared 

on the Web, and a brief span later examination and testing are finished by success specialists. In 

coordinate, a remote correspondence organize is produced that accomplices telephone with 

sensors or human body parts. Remedial issues are ending up more normal than later in continuous 

memory on the planet today. With the dynamic in helpful science now it has wound up being 

unquestionably not hard to see thriving related issues and the way of life of individuals which is 

finding the opportunity to be disgraceful legitimately. 

There are different foundations for fundamental remedial issues like unequal eating regimen, 

nonattendance of calories use, trademark issues, extend and whatnot. These have induced 

individuals coordinate with different wearisome honest to goodness illnesses like infection, 

diabetes, circulatory strain, thickness, and so on. Advantageous applications like Food Cam. 

Correspondence with Surrounding is utilized to screen person's dietary cases in worry with 

supplements. From this time forward thriving has changed into a basic key point for the people. 

The ascending of the correspondence advances has impelled different adjustments in the life of 

people. The web of things can comparatively serve the individual satisfaction of individuals 

experiencing various tribulations. This should be possible with the assistance of flourishing 

correspondence progresses 

D. Feedback Application 

Feedback adaptable application gives contribution about the effect of condition, for 

instance, essentialness usage, use of water and capacity to the customer. It goes about as a hover 

between the customer and the biological response. These applications respond a contribution to a 

customer. The ammeter structure screens or measures the essentialness usage of a home machine 

by a customer. It gives the contribution to the customer about its usage of essentialness by the 

family contraption with the objective that it can be checked or controlled. With this advancement, 

it is in like manner possible to finish an examination of the essentialness usage with online 

neighbors. Feedback application in PDAs furthermore measures the development and the rising

pollution which show the growing masses. This assistant in plotting some new benchmarks in 

remote identifying of the physical enveloping. 

E. Agriculture 

Robotization incredibly influences developing, making a sharp developing [16] method 

which allows the more viable developing. Adaptable applications interface the farmers with the 

plant hone which gives a fitting utilization of benefits. With flexible figuring application like 

MooMonitor +, agriculturists measure their cow-like prosperity and extravagance by putting 

sensors inside their body. With the compact applications, the farmer can offer their things by 

sharing them on the Web which allows the online courses of action. There are such countless 

supper applications through which farmers channel the things to check the idea of the sustenance. 

This can similarly check the expiry of the generation which prompts the diminishing of 

sustenance wastage. As agribusiness is surely not a shielded space, so adroit distinguishing and 

convenient preparing is imperative to save the item from bug, manage ideal pH and temperature. 

In this way, connecting versatile processing with IOT, including web data, permits quicker, 

appropriate more educated development of farming. 

F. Games and Sports 

These engaging exercises are the quickest developing applications in the field of Internet 

of Thing and Web of Thing innovation. In this classification, the portable application like Bike

Net, process the quickening, separate secured; breathing rate, and so forth of the sportsman. 

Additionally, there are a few sensors which put in shoes to figure the weight made by an athletic 

which encourages him to enhance his execution. Internet amusements and virtual diversions give 

the physical appearance of the portable client. Its collaboration with outside condition upgrades 

the gaming knowledge of the versatile client 

G. Interaction with Surroundings 

This is the more extensive class, which associates the surroundings with web-empowered 

administrations to upgrade or to mechanize the things. Straight to the point et al. exhibited the 

possibility of a looking framework in which sensor cell phones are utilized to locate the lost 

element. The cell phones are associated with advanced item code framework that peruses the 

RFID labels and standardized identification. 

The standardized tag examining or settling is conceivable through cell phones. The drug store 

therapeutic store is likewise focused by it, to advance the hunt procedure. The NFC-empowered 

cell phones are utilized to check the accessibility of a specific solution. With cell phone 

applications like My2cents clients would feedback be able to about the item and furthermore 

share the nature with the item. Associating with environment or things through versatile detecting 

requires different advancements like NFC, RFID, scanner tags and QR detecting. 

H. Transportation 

Web of Things or IOT aimed to connect transportation with the mobile application for 

better driving, easy parking and more convenience in public transport. For example, Densitybased 

traffic controller with IOT manages the traffic light system in accordance with the density 

of crowd on a particular junction and shared the information of traffic on the mobile application. 

Waze is the popular mobile application which connects 60 million users. It informs about the 

status of traffic and provides live-mapping, voice navigation system, alerts about road accidents 

and even more. INRIX traffic app provides similar services, offers parking services, based on the 

type of vehicle and the period of parking. 

V Track monitors the traffic delays by locating the driver’s phones. In, road situation 

monitoring and alert function are provided which offers the alerts about road condition to the 

user. Park here, is a sensing GPS parking facility which helps the driver in the automatic finding 

of vehicle parking. GoMetro is a mobile application which links location of the user with public

transport for convenience. In this field, authors like to do more research work for offering the 

better driving facility. The above mobile apps like GoMetro are getting popular especially in the 

USA. 

ANALYSIS AND OPEN PROBLEMS 

As above talked about, we can see the quantity of utilizations that join the Internet of 

Things or Web of Things with versatile detecting to get more educated and improved 

administrations. In this stage, we will do examination and talked about open issues related with 

individual application or work. With this study of related work, investigation has been shaped 

which are as per the following 

a) Cell phone sensors are utilized as a part of the participatory method to permit clients for the 

blend of information on the Web or Internet. 

b) The versatile applications are utilized to control gadgets (home mechanization) situated in the 

physical condition. 

c) The cell phones interface the body sensors with the web for games and wellbeing applications. 

d) The sensors are put inside the dairy animals body to gauge the wellbeing and fruitfulness of a 

cow. 

e) The portable application is utilized to check the status of activity and gives live-mapping. 

f) The portable application shares the data of various clients to encounter social cooperation. 

The detecting advances in every one of the eight applications are clarified. The information 

reconciliation from different cell phones is a costly undertaking and requires a ton of 

computation. As indicated by the overview, a few billions of dollars are utilized for these IOT 

works. It is normal that IOT profit will develop quickly in the coming years. Versatile figuring 

has an extraordinary effect in the field of wellbeing, transportation, control and Games or games. 

These applications have the most noteworthy business esteem as per the overview. Versatile 

applications in input and home mechanization are expanding. At last, the utilization of the 

versatile application in keen cultivating or agribusiness and association with things are still 

slower. There are a few difficulties, talked about in different reviews that ran over when portable 

detecting is joined with the Internet of Things or Web of Things, which are as per the following:

1. Regular Sensing 

A portion of the versatile applications require general detecting, information exchange through 

sensors which request high computational cost, stockpiling, and other equipment. There must be 

some other alternative in which sense is the second assignment in the cell phone. 

2. Mobile Crowd sensing 

It has different issues like how to quantify detecting quality, how to manage deficient information 

and how to utilize normal administrations. 

3. Security 

It is the most important in mobile sensing when interacting with IoT or WoT. Security must 

be ensuring while sharing the resources against unauthenticated users. High-security architecture 

like Yaler must be developed for the solution. 

4. Privacy 

Security is the significant issue or test for the client. Amid the sharing of individual data 

on the web, protection must be in thought. Versatile applications must be furnished with high 

security. 

5. Congestion 

While exchanging information or sharing assets, it must guarantee that the information 

being exchanged is sans blunder or exact. 

6. Precise 

It is imperative angle particularly in couple of uses like Health that comprise of body 

arrange. In this, the estimations must be dependable and exact. The versatile applications must 

recognize the imperatives. It additionally decides to what low-level of estimation can be 

performed. 

7. Personnel

It is normal that your cell phone alarms you about your wellbeing. Assume, you are going 

by your specialist and his telephone recommends him supplements and supplements for you, as 

per the status of your body sensors. 

8. Mood analysis 

It is practically equivalent to work force that offer better mentalities to the client to 

manage a circumstance as indicated by their feelings given by versatile application. It 

encourages representatives or laborers to enhance their wellbeing and productivity. 

9. Cloud storage 

It is fundamental in the vast majority of the group applications, keeping in mind the end 

goal to store enormous measure of information created by portable applications. Normal detecting 

prompts the age of enormous information which must be put or put away in an appropriate place. 

10. Market Cases 

It is imperative to get the ideal market idle of IOT or WOT with portable detecting, 

demonstrating the mechanized world. The financial attainability of IOT or Web-based resource 

must be clear. Every one of the issues talked about above influences the distinctive applications 

in an alternate way. 

As per the creators of numerous sorts of research, examined in this study, the result and 

criticalness of each issue are diverse in every application. They are separated as exceptionally 

basic, fundamental, critical, less vital and not applicable. For instance, Security is basic in 

wellbeing, control, input, less essential in swarm detecting, diversions, and games while vital in 

horticulture and communication with things. It is watched that security and protection are 

fundamental in the field of wellbeing, criticism, and transportation. Distributed storage is vital in 

the application where huge or enormous information is created like in swarm detecting. The 

answers for the current issues not just raise the use of versatile detecting in IOT or WOT yet 

additionally increment the market esteem particularly in the field of wellbeing, games, and 

transportation. Versatile application territory will be noteworthy or expanded in future 

incorporate shrewd city, tourism, seismic tremor and surge administration, in the safe exchange 

and small scale organizing inside the human body for wellbeing related data.

CONCLUSION 

This paper is an overview paper which portrays different utilizations of versatile 

detecting. The Internet of things and web advances have been a fundamental administration to 

make our lives exceptionally more straightforward furnishing us with the different applications in 

the cell phone which could detect the world through the advances. We have talked about the 

portable detecting and its applications in various fields. This is giving more dynamic data to the 

general population helping in better basic leadership in our life. We considered numerous 

examination papers and reasoned that the field of portable detecting is progressively getting to be 

huge and important. 

This paper is examination the eight distinct classifications of portable detecting (swarm detecting, 

criticism, wellbeing, transportation, sports, cooperation with the encompassing, agribusiness, 

control). Additionally, the difficulties of versatile detecting in various fields have been talked 

about. At long last, we have investigated the issues and proposed the particular arrangements 

enhance the utilization of versatile detecting all the more proficiently. Summing up, the web of 

things and web application has empowered us to exploit the new openings and the critical factor 

is to work in this field and grow more research to fulfill the difficulties

REFERENCES 

1. P. Sommer, L. Schor, and R. Wattenhofer, "Towards a zero arrangement remote sensor organize 

engineering for brilliant structures," in Proc. BuildSys, Berkeley, CA, USA, 2009, pp. 31-36. 

2. D. Yazar and A. Dunkels, "Proficient application reconciliation in IP-based sensor systems" in 

Proc.! st Workshop Embedded Sens. Syst. Vitality Efficiency Build. Berkeley, CA, USA, Nov. 

2009, pp. 43-48. 

3. J. W.Huiand D. E. Culler, "IP is dead, long live IP for remote sensor systems," in Proc. sixth 

Conf. Netw.Embedded Sensor Syst. (SenSys), Raleigh, NC, USA, Nov. 2008. 

4. E. Wilde, "Putting things to REST" School Inf., UC Berkeley, CA, USA, UCB iSchool Tech. 

Rep 2007-015, Nov. 2007. 

5. Kamilaris, A. Pitsillides, and M. Yiallouros, "Building vitality mindful shrewd homes utilizing 

Web innovations," J. Encompassing Intell. Savvy Environment, vol. 5, no. 2, 2013 

6. N.D. Path et al., "A Survey of cell phone detecting," IEEE Commun. Mag., vol. 5, no. 2, pp. 161- 

186, 2013. 

7. G. G. Meyer, K. Framling, and J. Holmstrom, "Canny items: An overview "Comput. Ind., vol. 

60, no.3, pp. 137-148, 2009.

SURVEY ON IMAGE PROCESSING METHODS USING THE FARMER 

PROFIT PRODUCTION 

M.Bhuvaneswari & R.Vinci Vinola 

Fatima College(Autonomous), Madurai 

ABSTRACT 

India is an agricultural country wherein about 70% of the population depends on 

agriculture. Agriculture has changed more in the past century than it has since farming began 

many millennia ago. Research in agriculture is aimed towards increase of productivity and food 

quality at reduced expenditure and with increased profit. Now day Computer technologies have 

been shown to improve agricultural productivity in a number of ways. One technique which is 

emerging as a useful tool is image processing. These techniques produce an output for the input 

image acquired through any of the imaging technique. The results are informative and supportive 

for farmers, agro based industries and marketers. It provides a timely support for decision making 

at an affordable cost. This paper presents a short survey on using image processing techniques to 

assist researchers and farmers to improve agricultural practices. 

Keywords 

Precision Agriculture; Image Processing; Profit for Farmer.

INTRODUCTION 

Now this year the continued demand for food with an increasing population, climate 

change and political instability. The agriculture industry continues to search for new ways to 

improve productivity and sustainability. This has resulted in researched from multiple disciplines 

searching for ways to incorporate new technologies and precision into the agronomic system. 

There is a need for efficient and precise techniques of farming, enabling farmers to put minimal 

inputs for high production. Precision agriculture is one of such techniques that is helping farmers 

in meeting both the above needs, precision agriculture can assist farmers in decision making 

about seed selection, crop production, disease monitoring, weed control, pesticide and fertilizers 

usage. It analyzes and controls farmer’s requirements using location specific information data and 

imagery techniques, Schellberg et al.(2008). In many parts of the world, Mainly in the rural areas 

this kind of data is inaccessible and the cost of procurement of these techniques is also not 

affordable by the farmers, Mondal and Basu(2009). The trends towards precision farming 

techniques are reliant on location specific data including the taking of multiple image databases. 

The use of image a survey of Image Processing Techniques for agriculture to be used for decision 

making 

IMAGE PROCESSING IN AGRICULTURAL CONTEXT 

Image processing techniques can be used to enhance agricultural practices by improving 

accuracy and consistency of processes while reducing farmers manual monitoring. Often it offers 

flexibility and effectively substitutes the farmer visual decision making. Image Processing term 

meaning image acquisition process of retrieving of a digital image from a physical source capture 

an image using sensors gray scale conversion process of converting. A color or multi channel 

digital image to a single channel where image pixel possess a single intensity value where image 

pixel possess a single intensity value image background, retrieving foreground objects Image 

enhancement improvement in perception of image details for human and machine analysis image 

histogram analysis pixel plot analysis in terms of peaks and valleys formed by pixel frequency 

and pixel intensities binary image segmentation foreground objects separation from background 

in a binary(black and white). 

Image color image segmentation image objects separation in color image, regions of interest 

image filtering process of distorting. An image in a desired was using a filter feature extraction 

process of defining a set of features or image characteristics that efficiently or meaningfully 

represent the information important for analysis and classification image registration process of

transforming different sets of data into one coordinate system image transition process of 

transforming different sets of data into one co ordinate system image transition process of 

changing state or defining a condition between two or more images image object detection 

process of finding instances of real world objects such as weeds, plants and insects in images or 

video sequences image object analysis process extracting reliable and meaningful information 

from images. Table (1) Applications/ Models/ Systems Developed Using Image Processing 

Techniques And Remarks Over Its Accuracy And Usability. 

REFERENCES 

OBJECTIVE OF 

METHOD 

RESULT 

THE STUDY 

Payne Et Al.,2014 

To Predict Mango 

Image Processing 

Mango Fruits Were Detected With 

Yield 

78.3% 

Prathibha Et Al., 

2014 

Early Pest Detection Image Processing System Can Detect The Pests In 

Tomato Fruit In The Early Stage. 

Intaravanne & 

To Predict The 

Android Device 

Leaf Color Levels Can Be 

Sumriddetch, 2015 

Needed Fertilizer For 

And Image 

Accurately Identified 

Rice 

Processing 

Mainkar Et Al.2015 To Automatically 

Detect Plant Leaf 

Diseases 

Image Analysis 

Leaf Diseases Were Detected By 

The Proposed System 

Gopal, 2016 

To Develop An Auto 

Image Processing 

Developed Model Was Useful For 

Irrigation And Pest 

Irrigation And Pest Detection 

Detection System 

Hanson Et Al.2016 

For Detection And 

Image Processing, 

Watermelon Leaf Diseases Were 

Identification Of 

Neural Networks 

Detected With 75.9% Of 

Plant Disease 

Accuracy 

Zhao Et Al., 2016 

To Predict Oriental 

Digital Image 

Developed System Was 

Fruit Moths 

Processing 

Successful For Prediction (0.99 

Technology 

,P

Maharlooei Et Al., To Detect And Count E-Image Image Captured With An 

2017 Different Sized Processing Inexpensive Digital Camera Gave 

Soybean Aphids On Technique Satisfactory Results 

A Soybean Leaf 

Gray Scale Conversion 

After image acquisition, pre processing of the images involves gray scale conversion, 

Eerens et al.(2014) and Jaya et al.(2000). Du and Sun(2004) highlights gray scale conversion as 

an intermediate step in food quality evaluation models. They reported various applications 

evaluating food items like fruits, fishery, grains, meat, vegetables and others the use of image 

processing techniques applicable to different assessments. Other work (2013) reported on the use 

of gray level determination of foreign fibers in cotton production images that enhanced 

background separation and segmentation. Java et al.(2000) also demonstrated the image analysis 

techniques using neural networks for classification of the agriculture products. This study 

reported that the multi-layer neural networks classifiers are the best in performing categorization 

of agricultural products. For example here: 

Image Background Extraction in Applications 

The Background is of minimal use it is preferable to extract if from the images. Such 

images having regions of interests solid objects in dissimilar background are easily extractable. 

This results in non-uniform gray levels distribution between objects of interests and the image 

background, Eerens et al.(2014) and Java et al.(2000) following this understanding du and 

sun(2004) report various applications where background is not taken into consideration while

evaluating the food products quality including pizza, corn germplasm and cob,etc. Similarly wu 

et al.(2013) extracted background of the foreign fiber images detected in to cotton products. This 

aids in the clear detection of foreign fibers which were difficult to trace out. A survey on 

advanced techniques by Sankaran et al.(2010) highlight the us of fluorescence spectroscopy and 

imaging, visible and infrared spectroscopy, hyper spectral imaging in detecting plant diseases and 

on future enhancements, which could focus on the metabolic activities of the plants and trees 

releasing volatile organic compounds. 

DISCUSSION 

The review of survey papers on the use of image processing techniques showed that these 

techniques can be useful to assist agricultural scientists. The deep approach which is improve 

application in computer vision, automatic speech recognition and natural language processing, 

Bengio(2009) is emerging as the preferred approach. The review found that crop identification 

and disease detection the common uses for the technique. 

CONCLUSION 

This paper presented a survey on using image processing techniques used in an agriculture 

context. Employing the processes like segmentation, features extraction and clustering can be 

used to interrogate images of the crops. There is a need to select the most appropriate techniques 

to assist decision-making. The image processing techniques have been used across a vast range of 

agricultural production contexts. It can be effective in food quality assessment, fruit defects 

detection, weed crop classification. There are a number of applications and methods to choose 

from for implementation to real time needs. While the existing application sustaining the needs of 

today, there are more and more new methods are evolving to assist and ease the farming 

practices.It is evident that these approaches will all contribute to the wider goal of optimizing 

global of production. One factor, which could increase the development of image processing 

techniques for agriculture is the availability of Online data sets. No online images databases are 

available on food quality assessment, fruit defects detection or weed/crop classification. Similar

to databases like handwritten or printed documents and characters, faces, there is a need of 

agricultural databases that will ease in the testing and verification of newly developed image 

processing methods. 

REFERENCES 

1. Lalit P. Saxena1 and Leisa J.(2014) Armstrong2, A survey of image processing techniques for 

agriculture. 

2. A.B. Ehsanirad And Y. H. S. Kumar (2010), Leaf Recognition For Plant Classification Using 

Glcm And Pca Methods, Oriental Journal Of Computer Science & Technology, 3 (1), Pp. 31–36, 

2010. 

3. Image Analysis Applications In Plant Growth And Health Assessment, Journal Of Agricultural 

Faculty Of Mustafa Kemal University.(2017). 

4. C. C. Yang, S. O. Prasher, J. A. Landry, J. Perret And H. S. Ramaswamy (2000), Recognition Of 

Weeds With Image Processing And Their Use With Fuzzy Logic For Precision Farming, 

Canadian Agricultural Engineering, 42 (4), Pp. 195-200, 2000. 

5. D. S. Jayas, J. Paliwal and N. S. Visen (2000), Multi-layer neural networks for image analysis of 

agricultural products, J. Agric. Engng Res., 77 (2), pp. 119–128, 2000. 

6. D. J. Mulla (2013), Twenty five years of remote sensing in precision agriculture: Key advances 

and remaining knowledge gaps, Biosystems Engineering, 114, pp. 358–371, 2013. 

7. E. E. Kelman and R. Linker (2014), Vision-based localisation of mature apples in tree images 

using convexity, Biosystems Engineering, 114, pp. 174–185, 2014. 

8. C. J. Du and D. W. Sun (2004), Recent developments in the applications of image processing 

techniques for food quality evaluation, Trends in Food Science & Technology, 15, pp. 230–249, 

2004. 

9. C. Puchalski, J. Gorzelany, G. Zagula and G. Brusewitz (2008), Image analysis for apple defect 

detection, detection, Biosystems and Agricultura Engineering, 8, pp. 197–205, 2008.

AN OVERVIEW OF CLOUD COMPUTING 

R. Thirumalai Kumar& V.Vinoth Kumar 

NMSSVN College,Madurai 

ABSTRACT 

The development of Cloud Computing services is speeding up the rate in which the 

organizations outsource their computational services or sell their idle computational resources. 

Even though migrating to the cloud remains a tempting trend from a financial perspective, there 

are several other aspects that must be taken into account by companies before they decide to do 

so cloud computing has become a key IT buzzword. Cloud computing is in its infancy in terms of 

market adoption. However, it is a key IT megatrend that will take root. Aiming to give a better 

understanding of this complex scenario, in this paper giving an overview of the cloud computing 

in this emerging technology. 

INTRODUCTION 

Cloud computing is a subscription-based service where you can obtain networked storage 

space and computer resources. Cloud computing entails running computer/network applications 

that are on other people’s servers using a simple user interface or application format. The cloud 

technology has many benefits and that would explain its popularity. First, companies can save a 

lot of money; second, they are able to avoid the mishaps of the regular server protocols. When a 

company decides to have a new piece of software, whose license can only be used once and it’s 

pretty expensive, they wouldn’t have to buy software for each new computer that is added to the 

network. Instead, they could use the application installed on a virtual server 

somewhere and share, in the ‘cloud’.

How can we use the cloud? 

The cloud makes it possible for you to access your information from anywhere at any 

time. While a traditional computer setup requires you to be in the same location as your data 

storage device, the cloud takes away that step. The cloud removes the need for you to be in the 

same physical location as the hardware that stores your data. Your cloud provider can both own 

and house the hardware and software necessary to run your home or business applications. 

This is especially helpful for businesses that cannot afford the same amount of hardware and 

storage space as a bigger company. Small companies can store their information in the cloud, 

removing the cost of purchasing and storing memory devices. Additionally, because you only 

need to buy the amount of storage space you will use, a business can purchase more space or 

reduce their subscription as their business grows or as they find they need less storage space. 

One requirement is that you need to have an internet connection in order to access the cloud. This 

means that if you want to look at a specific document you have housed in the cloud, you must 

first establish an internet connection either through a wireless or wired internet or a mobile 

broadband connection. The benefit is that you can access that same document from wherever you 

are with any device that can access the internet. These devices could be a desktop, laptop, tablet, 

or phone. This can also help your business to function more smoothly because anyone who can 

connect to the internet and your cloud can work on documents, access software, and store data. 

Imagine picking up your smart phone and downloading a pdf document to review instead of 

having to stop by the office to print it or upload it to your laptop. This is the freedom that the 

cloud can provide for you or your organization.

CHARACTERISTICS OF THE CLOUD 

‣ The “no-need-to-know” in terms of the underlying details of infrastructure, applications 

interface with the infrastructure via the APIs. 

‣ The “flexibility and elasticity” allows these systems to scale up and down at will 

utilizing the resources of all kinds (CPU, storage, server capacity, load balancing, and databases) 

‣ The “pay as much as used and needed” type of utility computing and the “always on 

anywhere and any place” type of network-based computing. 

• Cloud are transparent to users and applications, they can be built in multiple ways 

‣ Branded products, proprietary open source, hardware or software, or just off-the-shelf 

PCs. 

In general, they are built on clusters of PC servers and off-the-shelf components plus Open 

Source software combined with in-house applications and/or system software. 

TYPES OF CLOUDS 

There are different types of clouds that you can subscribe to depending on your needs. As 

a home user or small business owner, you will most likely use public cloud services.

1. Public Cloud - A public cloud can be accessed by any subscriber with an internet 

connection and access to the cloud space. 

2. Private Cloud - A private cloud is established for a specific group or organization and 

limits access to just that group. 

3. Hybrid Cloud - A hybrid cloud is essentially a combination of at least two clouds, 

where the clouds included are a mixture of public and private. 

CHOOSING A CLOUD PROVIDER 

Each provider serves a specific function, giving users more or less control over their cloud 

depending on the type. When you choose a provider, compare your needs to the cloud services 

available. Your cloud needs will vary depending on how you intend to use the space and 

resources associated with the cloud. If it will be for personal home use, you will need a different 

cloud type and provider than if you will be using the cloud for business. Keep in mind that your 

cloud provider will be pay-as-you-go, meaning that if your technological needs change at any 

point you can purchase more storage space (or less for that matter) from your cloud provider. 

There are three types of cloud providers that you can subscribe to: Software as a Service 

(SaaS), Platform as a Service (PaaS), and Infrastructure as a Service (IaaS). These three types

differ in the amount of control that you have over your information, and conversely, how much 

we can expect our provider to do for. Briefly, here is what can expect from each type. 

1. Software as a Service - A SaaS provider gives subscribers access to both resources and 

applications. SaaS makes it unnecessary for you to have a physical copy of software to install on 

your devices. SaaS also makes it easier to have the same software on all of your devices at 

once by accessing it on the cloud. In a SaaS agreement, you have the least control over the cloud. 

2. Platform as a Service - A PaaS system goes a level above the Software as a Service 

setup.APaaS provider gives subscribers access to the components that they require to develop and 

operate applications over the internet. 

3. Infrastructure as a Service - An IaaS agreement, as the name states, deals primarily with 

computational infrastructure. In an IaaS agreement, the subscriber completely outsources the 

storage and resources, such as hardware and software, that they need. 

OPPORTUNITIES AND CHALLENGES OF CLOUD COMPUTING 

 

 

 

The use of the cloud provides a number of opportunities 

It enables services to be used without any understanding of their infrastructure. 

Cloud computing works using economies of scale: 

It potentially lowers the outlay expense for start up companies, as they would no 

longer need to buy their own software or servers. 

Cost would be by on-demand pricing. 

Vendors and Service providers claim costs by establishing an ongoing revenue 

stream. 

Data and services are stored remotely but accessible from “anywhere”. 

 

In parallel there has been backlash against cloud computing

Use of cloud computing means dependence on others and that could possibly limit 

flexibility and innovation: 

The others are likely become the bigger Internet companies like Google and IBM, 

who may monopolise the market. 

Some argue that this use of supercomputers is a return to the time of mainframe 

computing that the PC was a reaction against. 

 

Security could prove to be a big issue: 

 

It is still unclear how safe out-sourced data is and when using these services 

ownership of data is not always clear. 

 

 

 

 

 

data. 

There are also issues relating to policy and access: 

If your data is stored abroad whose policy do you adhere to? 

What happens if the remote server goes down? 

How will you then access files? 

There have been cases of users being locked out of accounts and losing access to 

ADVANTAGES OF CLOUD COMPUTING 

Lower computer costs 

o You do not need a high-powered and high-priced computer to run cloud 

computing's web-based applications. 

o Since applications run in the cloud, not on the desktop PC, your desktop PC does 

not need the processing power or hard disk space demanded by traditional desktop 

software. 

o When you are using web-based applications, your PC can be less expensive, with a 

smaller hard disk, less memory, more efficient processor... 

o In fact, your PC in this scenario does not even need a CD or DVD drive, as no 

software programs have to be loaded and no document files need to be saved.

Improved performance 

o With few large programs hogging your computer's memory, you will see better 

performance from your PC. 

o Computers in a cloud computing system boot and run faster because they have 

fewer programs and processes loaded into memory… 

Reduced software costs 

o Instead of purchasing expensive software applications, you can get most of what 

you need for free-ish 

• applications today, such as the Google Docs suite. 

o better than paying for similar commercial software 

• which alone may be justification for switching to cloud applications. 

Instant software updates 

o Another advantage to cloud computing is that you are no longer faced with 

choosing between obsolete software and high upgrade costs. 

o When the application is web-based, updates happen automatically 

• available the next time you log into the cloud. 

o When you access a web-based application, you get the latest version 

• without needing to pay for or download an upgrade. 

Improved document format compatibility 

o You do not have to worry about the documents you create on your machine being 

compatible with other users' applications or OSes 

o There are potentially no format incompatibilities when everyone is sharing 

documents and applications in the cloud.

Unlimited storage capacity 

o Cloud computing offers virtually limitless storage. 

o Your computer's current 1 Tbyte hard drive is small compared to the hundreds of 

Tbytes available in the cloud. 

Increased data reliability 

o Unlike desktop computing, in which if a hard disk crashes and destroy all your 

valuable data, a computer crashing in the cloud should not affect the storage of 

your data. 

• if your personal computer crashes, all your data is still out there in the 

cloud, still accessible 

o In a world where few individual desktop PC users back up their data on a regular 

basis, cloud computing is a data-safe computing platform! 

Universal document access 

o That is not a problem with cloud computing, because you do not take your 

documents with you. 

o Instead, they stay in the cloud, and you can access them whenever you have a 

computer and an Internet connection 

o Documents are instantly available from wherever you are 

Latest version availability 

o When you edit a document at home, that edited version is what you see when you 

access the document at work. 

o The cloud always hosts the latest version of your documents 

o as long as you are connected, you are not in danger of having an outdated version 

Easier group collaboration

o Sharing documents leads directly to better collaboration. 

o Many users do this as it is an important advantages of cloud computing 

o Multiple users can collaborate easily on documents and projects 

Device independence 

o You are no longer tethered to a single computer or network. 

o Changes to computers, applications and documents follow you through the cloud. 

o Move to a portable device, and your applications and documents are still available. 

DISADVANTAGES OF CLOUD COMPUTING 

Requires a constant Internet connection 

o Cloud computing is impossible if you cannot connect to the Internet. 

o Since you use the Internet to connect to both your applications and documents, if 

you do not have an Internet connection you cannot access anything, even your 

own documents. 

o A dead Internet connection means no work and in areas where Internet 

connections are few or inherently unreliable, this could be a deal-breaker. 

Does not work well with low-speed connections 

o Similarly, a low-speed Internet connection, such as that found with dial-up 

services, makes cloud computing painful at best and often impossible. 

o Web-based applications require a lot of bandwidth to download, as do large 

documents. 

Features might be limited 

o This situation is bound to change, but today many web-based applications simply 

are not as full-featured as their desktop-based applications. 

• For example, you can do a lot more with Microsoft PowerPoint than with 

Google Presentation's web-based offering

Can be slow 

o Even with a fast connection, web-based applications can sometimes be slower than 

accessing a similar software program on your desktop PC. 

o Everything about the program, from the interface to the current document, has to 

be sent back and forth from your computer to the computers in the cloud. 

o If the cloud servers happen to be backed up at that moment, or if the Internet is 

having a slow day, you would not get the instantaneous access you might expect 

from desktop applications. 

Stored data might not be secure 

o With cloud computing, all your data is stored on the cloud. 

• The questions is How secure is the cloud? 

o Can unauthorised users gain access to your confidential data? 

Stored data can be lost 

o Theoretically, data stored in the cloud is safe, replicated across multiple machines. 

o But on the off chance that your data goes missing, you have no physical or local 

backup. 

• Put simply, relying on the cloud puts you at risk if the cloud lets you down. 

THE FUTURE 

• Many of the activities loosely grouped together under cloud computing have already been 

happening and centralised computing activity is not a new phenomena 

• Grid Computing was the last research-led centralised approach 

• However there are concerns that the mainstream adoption of cloud computing could cause 

many problems for users 

• Many new open source systems appearing that you can install and run on your local cluster 

– It should be able to run a variety of applications on these systems

CONCLUSION 

To summarize, the cloud provides many options for the everyday computer user as well as 

large and small businesses. It opens up the world of computing to a broader range of uses and 

increases the ease of use by giving access through any internet connection. However, with this 

increased ease also come drawbacks. You have less control over who has access to your 

information and little to no knowledge of where it is stored. You also must be aware of the 

security risks of having data stored on the cloud. The cloud is a big target for malicious 

individuals and may have disadvantages because it can be accessed through an unsecured internet 

connection. If you are considering using the cloud, be certain that you identify what information 

you will be putting out in the cloud, who will have access to that information, and what you will 

need to make sure it is protected. Additionally, know your options in terms of what type of cloud 

will be best for your needs, what type of provider will be most useful to you, and what the 

reputation and responsibilities of the providers you are considering are before you sign up. 

REFERENCE 

http://csrc.nist.gov/publications/drafts/800-145/Draft-SP-800-145_cloud-definition.pdf. 

Won Kim: “Cloud Computing: “Status and Prognosis”, in Journal of Object Technology, vol. 8, 

no.1, January-February 2009, pp. 65-72 

http://www.jot.fm/ Cloud Computing: Today and Tomorrow

A STUDY ON IMAGE PROCESSING IN AGRICULTURE TO DETECT 

INTRODUCTION 

PLANT DISEASE 

A.L.Deepika,S.Vinothini 

Fatima College(Autonomous),Madurai 

Image processing is a method to perform some operations on an image, in order to get an 

enhanced image or to extract some useful information from it. It is a type of signal processing in 

which input is an image and output may be image or characteristics/features associated with that 

image. Nowadays, image processing is among rapidly growing technologies. It forms core 

research area within engineering and computer science. 

Image processing basically includes the following three steps: 

1. Importing the image via image acquisition tools 

2. Analyzing and manipulating the image 

3. Output in which result can be altered image or report that is based on image analysis. 

There are two types of methods used for image processing namely, analog and digital 

image processing. Analog image processing can be used for the hard copies like printouts and 

photographs. Image analysts use various fundamentals ofinterpretation while using these visual 

techniques. Digital image processing techniques help in manipulation of the digital images by 

using computers. The three general phases that all types of data have to undergo while using 

digital technique are pre-processing, enhancement, and display, information extraction. DIP is the 

use of computer algorithms to create, process, communicate, and display digital images. The 

input of that system is a digital image and the system process that image using efficient 

algorithms, and gives an image as an output. The process of digital image processing is defined in 

the form of phases.

Image Acquisition 

Image Pre- Processing 

Image Segmentation 

Feature Extraction 

Classification based on 

Statistical Analysis 

DETECTING THE PLANT DISEASE 

In recent years the importance of sustainable agriculture has risen to become one of the most 

important issues in agriculture. In addition, plant diseases continue to play a major limiting role in 

agricultural production. The control of plant diseases using classical pesticides raises serious concerns 

about food safety, environmental quality and pesticide resistance, which have dictated the need for 

alternative pest management techniques. In particular, nutrients could affect the disease tolerance or 

resistance of plants to pathogens. When some diseases are not visible to naked eye but actually they 

are present, then it is difficult to detect it with the naked eye. Earlier, microscope is used to detect the 

disease, but it become difficult as to observe each and every leaf and plant. So, the fast and effective 

way is a remote sensing technique. Detection and recognition of diseases in plants using machine 

learning is very fruitful in providing symptoms of identifying diseases at its earliest. Computer 

processing Systems are developed for agricultural applications, such as detection of leaf diseases, 

fruits diseases etc. In all these techniques, digital images are collected using a digital camera and 

image processing techniques are applied on these images to extract useful information that are 

necessary for further analysis. Digital Image processing is used for the implementation which will 

take the image as input and then perform some operation on it and then give us the required or 

expected output.

The image processing can be used in agricultural applications for following purposes: 

1. To detect diseased leaf, stem, fruit 

2. To quantify affected area by disease. 

3. To find shape of affected area. 

4. To determine color of affected area 

5. To determine size & shape of fruits 

IMAGE ACQUISITION 

The image acquisition is required to collect the actual source image. An image must be 

converted to numerical form before processing. This conversion process is called digitization. 

IMAGE PREPROCESSING 

The principle objective of the image enhancement is to process an image for a specific task 

so that the processed image is better viewed than the original image. Image enhancement 

methods basically fall into two domains, spatial and frequency domain. 

IMAGE SEGMENTATION 

In image processing, segmentation falls in to the category of extracting different image 

attributes of an original image. 

Segmentation subdivides an image into constituent regions or objects 

CLASSIFICATION TECHNIQUE 

There are three important techniques are used. They are 

1. Artificial Neural Network 

2. Clustering Method 

3. SVM (Support Vector Machine

ARTIFICIAL NEURAL NETWORK 

Description 

ANN means Artificial Neutral Network. It is a type of artificial intelligence that includes some 

functions of the person mind. 

ANN’s have three layers that are interconnected. 

1. The first layer consists of input neurons. Those neurons send data on to the second layer, 

which in turn sends the output neurons to the third layer. 

2. ANN is also known as a neural network. 

Characteristics 

A large number of very simple processing neuron-like processing elements . A large number of 

weighted connections between the elements . Distributed representation of Knowledge over the 

connections . Knowledge is acquired by network through a learning process. 

Advantages 

1. It is a non-parametric classifier. 

2. It is an universal functional with arbitrary accuracy. 

3. capable to present functions such as OR, AND, NOT 

4. It is a data driven self-adaptive technique 

5. Efficiently handle. noisy inputs 

Disadvantages 

1. It is semantically poor. 

2. ANN is time taking. 

3. Problem of over fitting. 

CLUSTERING METHOD 

Description 

This is an iterative technique that is used to partition an image into clusters. Clusters can be 

selected manually or randomly. Distance between the pixel and cluster center is calculated by the 

squared or absolute. The difference is typically based on pixel color, intensity, texture, and location, 

or a weighted combination of these factors. More commonly used clustering algorithm 

k- means algorithm, fuzzy c-means algorithm, expectation – maximization (EM) algorithm.


Clustering based on intensity and threshold separation. 

1. It uses stochastic approach 

2. Performance and accuracy depends upon the threshold selection. 

Advantages 

1. Simpler classifier as exclusion of any training process. 

2. Applicable in case of a small dataset which is not trained 

Disadvantages 

1. Speed of computing distance increases according to numbers available in training 

samples. 

2. Expensive testing of each instance and sensitive to irrelevant inputs 

SVM (Support Vector Machine) 

Description 

A support vector machine builds a hyper plane or set of hyper planes in a high- or infinite 

dimensional space, used for classification, generalization error of the classifier. 


SVM uses Nonparametric with binary classifier approach and can handle more input data very 

efficiently. Performance and accuracy depends upon the hyper plane selection and kernel 

parameter. 

Advantages 

1. It gains flexibility in the choice of the form of the threshold. 

2. Contains a nonlinear transformation. 

3. It provides a good Generalization capability. 

4. The problem of over fitting is eliminated. 

5. Reduction in computational complexity.

Disadvantages 

1. Result transparency is low. 

2. Training is time consuming. 

3. Structure of algorithm is difficult to understand. 

CONCLUSION 

Therefore, this system will help the farmer to increase the production of agriculture. This 

method we can identify the disease present in monocot or dicot. Most of us we mainly used 

popular method, clustering method. This deals with the accuracy and fast approach of disease 

detection. By using this concept the disease identification is done for all kinds of plants and also 

the user can know the affected area of the plants in percentage. By identifying the disease 

properly, the user can rectify the problem very easy and with less cost. So I conclude that image 

processing is one of the important tools for disease detection of plants. 

REFERENCES 

1. P.Revathi, M.Hemalatha, “Classification of Cotton Leaf Spot Diseases Using Image 

Processing Edge Detection Technique”, ISBN, 2012, 169-173, IEEE. 

2. Santanu Phadikar & Jaya Sil[2008] Rice Disease Identification Using Pattern Recognition 

Techniques, 

3. Agrios, George N. (1972). Plant Pathology (3rd ed.). Academic Press.

ABSTRACT 

INTERNET OF BIOMETRIC THINGS IN FINGERPRINT 

TECHNOLOGIES AND DATABASES 

S.Jaishreekeerthana & S.Kowsalya 

MAM College,Trichy 

Since security vulnerabilities speak to one of the great difficulties of the Internet of 

Things, scientists have proposed what is known as the Internet of Biometric Things, which 

blends customary biometric advancements with setting mindful validation procedures. A 

standout amongst the most popular biometric innovations is electronic fingerprint 

acknowledgment, which get fingerprint utilizing diverse advancements, some of which are 

more appropriate for others. In addition, diverse fingerprint databases have been worked to 

contemplate the effect of different factors on the exactness of various fingerprint 

acknowledgment calculations and frameworks. In this paper, we study the accessible 

fingerprint obtaining advances and the accessible fingerprint databases. We likewise 

distinguish the preferences and disservices of every innovation and database. List Terms— 

Capacitive, Digital Cameras, Fingerprints, Internet of BiometricThings, Optical, Synthetic 

Fingerprints Generation, Thermal, Ultrasonic. 

INTRODUCTION 

Later, billions of items (or things) will be associated together making what is known as the 

Internet of Things .security vulnerabilities speak to one of the fabulous difficulties Utilizing 

biometric identification strategies to address a portion of the Security vulnerabilities is an 

exceptionally engaging arrangement. As of late, a few scientists have proposed what is known 

as the Internet of Biometric Things , which makes utilization of customary biometric 

identification strategies and setting mindful verification systems .A standout amongst the most 

popular biometrics innovations is electronic Fingerprint verification. It can be utilized inside 

various applications, for example, cell phones, ATMs, shrewd homes, brilliant structures, 

transportation frameworks, and so on. Along these lines, computerized fingerprint 

identification is getting consistently developing enthusiasm from both scholarly and business 

parties. The significance of concentrating on fingerprint, as a principle biometric device to 

exceptionally recognize individuals and henceforth the security and protection comes from 

numerous reasons. The first one is that no two people Supported by the Jordan University of 

Science and Technology Deanship of Research

Fig. 1: A Fingerprint Clarification have been found to have the same fingerprints. Another 

reason is that fingerprints don't change with age. At long last, each finger of a solitary 

individual has an alternate fingerprint. This reality can be used to distinguish not just the 

individual who contacted a specific question, yet in addition the finger utilized by that 

individual. As Shown in Figure 1, a fingerprint is framed from the accompanying parts: Ridge 

Endings, the terminals of the edges. Bifurcations, the purpose of separating edge into two 

branches. Specks,little edges. Islands irregular point without any connectors. Pore: a white 

point encompassed by dark point. Scaffolds, little edges joining two longer contiguous edges 

Crossovers: two edges which cross each other Core: focal point of the fingerprint. Deltas: 

focuses, at the lower some portion of the fingerprint, encompassed by a triangle. Fingerprints 

can be classified as per their shapes into three fundamental writes; Whorl fingerprint, Arch 

fingerprint and circle fingerprint. Each class has its edge flow and design. Figure 2 

demonstrates these three sorts with their reality rates. Circle fingerprint are the most wellknown 

with a percent (60% 65%) of all fingerprints. 

Curve fingerprints which are once in a while exist represents just 5% percent[6]. An average 

Automatic Fingerprint Identification System (AFIS) more often than not comprises of 

numerous phases that incorporate Image Obtaining, fingerprint division,image preprocessing 

(improvement), highlight extraction, and coordinating outcome . 

• Image Acquisitio 

The phase of picture procurement is a standout amongst the most essential factors that 

contribute in the achievement of the fingerprint acknowledgment framework, here there are a 

few elements to be considered before picking the gadget that will be utilized to catch 

fingerprints regarding execution, cost and size of the subsequent picture, in this investigation 

we center around the fingerprint catch stage as it were.• Segmentation: The principle of this 

step is to separate the image of the finger from the background, it is an initial step of 

fingerprint system

• Preprocessing 

This phase includes some modifications to the image with the aim of enhancement, this would 

increase the accuracy of the system used in fingerprint recognition.. 

• Feature extraction and Person 

At last, comparable fingerprints are recovered. Along these lines, just this arrangement 

of fingerprints (source) are contrasted and the objective fingerprint. Fingerprints are gained in 

various ways. For non-mechanized (or non-on the web) frameworks, fingerprints are typically 

procured by stamping inked fingertips on an extraordinary paper. On the off chance that these 

fingerprints should be put away on a computerized gadget, a picture of the paper is taken 

utilizing a scanner for instance. Then again, fingerprints are procured utilizing fingertips 

scanners for mechanized frameworks. Scanners may utilize distinctive advances, for example, 

optical, capacitive, warm, and ultra-sonic. These scanners more often than not require the 

fingertip to be squeezed or hauled over some surface. This may reshape or crush the 

subsequent fingerprint which may prompt mistaken identification. Likewise, this makes it 

difficult to cross utilize diverse finger tips scanners in securing and verification process as 

various scanners may brings about various reshaping comes about. As of late, as most present 

day computerized frameworks are given, or can be effortlessly given, with fantastic advanced 

cameras, a course of research began to consider.The likelihood of securing fingerprints 

utilizing these cameras. As indicated by the investigations in this approach is promising and 

the outcomes are worthy. To plan distinctive procedures and concentrate the effect of various 

components identified with robotized fingerprint acknowledgment frameworks, an expansive 

fingerprint picture set is required .In this dad per, we go for cultivating the improvement of 

concentrating on fingerprint verification. Specifically, we direct a review on the accessible 

fingerprint picture sets and the procurement gadgets used to gather each, for example, optical 

sensor, capacitive sensor, warm sensor, and diverse kinds of advanced cameras. We list the 

fundamental highlights of every innovation and fingerprint picture set, next to their favorable 

circumstances and hindrances.

FINGERPRINT ACQUISITION DEVICES 

This section gives an overview of the most common devices used for online fingerprint 

acquisition. Many technologies have been utilized in these devices. Namely, optical, 

capacitive, thermal, and ultrasonic 

A. Optical sensor 

Optical fingerprint devices are the oldest and the most commonly used devices to 

capture fingerprints. Figure 3(a) 1 clarifies how this sensor works. The fingerprint image is 

captured by placing a finger on one side of a prism, a light source that emits light vertical on 

the second prism side, and a camera that is placed in front of the third prism side. The light 

collected by the camera is then converted into a digital signal where bright areas reflect the 

valleys and dark areas reflect the ridges.This sensor was used to acquire a large number finger 

print databases, such as FVC2000 (DB1, DB3), FVC2002 (DB1 and DB2), FVC2004 (DB1 

and DB2), FVC2006 (DB2). Figure 3(b) is an example of a commercial optical sensor. The 

main advantages of optical sensors are temperature insensitivity, low cost, and high resulting 

fingerprint resolution. On the other hand, it suffers form many disadvantages. As optical 

sensor depends on light reflection, as a result, its functionality may suffer from lighting 

environment. 

A) Clarification of how optical sensors work

B) Commercial optical sensor based Fingerprint 

Acquisition Device picture as they may cover. Besides, this deposit can be utilized 

to take the fingerprint. One of the business cases that uses this innovation is appeared in 

Figure 3(b) is the SecuGen Hamster IV 2. it is one of the normal gadgets in fingerprint 

frameworks, which have been certified by the FBI and STQC. Close to SecuGen Hamster IV, 

other optical gadgets were utilized to gather the IIIT-D Multi-sensor Optical and Latent 

Fingerprint .B. Capacitive Sensor 

This gadget depends on the standard of estimating the electric limit of the human skin. 

Fingerprints are set on a thin plate that contains an arrangement of small scale capacitors. Skin 

of a fingerprint contains edges and valleys. The capacitive sensor fundamentally measure the 

limit of the skin of the finger. The limit of edges is diverse frame the limit of valleys on the 

grounds that the limit depends very on the separation of the skin from the miniaturized scale 

capacitor plate. A basic model of a capacitive sensor is appeared. This sensor was utilized to 

catch various fingerprint datasets that incorporates FVC2000 (DB2) [20] and FVC2002 (DB3) 

. Favorable position of capacitive sensors are little in estimate and consequently they are less 

expensive to deliver. They additionally devours Capacitive Technology low energy.This 

advantage is especially helpful in the battery controlled gadgets. It likewise obtuse to 

ecological impacts, for example, daylight. Numerous producers, for example, Sony, Fujitsu, 

Hitachi, and Apple, have embraced this innovation in their fingerprint acknowledgment 

frameworks.C. Synthetic Fingerprint Generation

This strategy naturally creates artificial fingerprints in light of a model. This model 

typically is controlled by an arrangement of parameters which can be changed to deliver 

variant fingerprints as far as clearness, territory, focused or not, revolution, the difficulty of 

the acknowledgment of the fingerprint, and some more. This technique can be utilized to 

manufacture substantial fingerprint databases in brief time with no cost. Additionally, the 

produced artificial fingerprints might be tuned to recreate normal fingerprints. This technique 

was utilized as a part of some fingerprint databases, Sensor depends on pyro electric material. 

This material can change temperature contrasts to various voltages. By the contrast between 

the temperatures of the edges and the valleys, the voltage created for every one is not the same 

as the other. 

A standout amongst the most widely recognized warm sensors is "Atmel FingerChip" 

which is utilized to secure numerous outstanding fingerprint databases.Namely, FVC2004 

(DB3) and FVC2006 (DB3). This technology is not very common commercially. One of the 

difficulties faced when dealing with this technology is that the thermal image should be 

captured in a very short time. That is because when a finger is placed over the pyro electric 

material, the temperature difference between the areas under the ridges and the valleys will be 

initially detectable. Eventually, after a short period of time, the temperature difference 

vanishes as the adjacent areas heat each other. As this method depends on body temperature, it 

is very hard to be received E. Ultrasound Sensor 

This gadget depends on the rule of sending high recurrence sound waves, and after that 

catching the reflection from the finger surface. The upside of this innovation is its capacity to 

go through residue, soil, and skin oil that may cover the finger surface. Therefore, fingerprint 

ultrasound detecting might be utilized to catch top notch pictures if the finger is messy. Then 

again, this method has numerous weakness. For instance, it is exceptionally costly and may 

set aside long opportunity to catch the fingerprint picture. Qualcomm4 are using ultrasonic 

innovation

a) Clarification of how Ultrasonic Sensors work 

F. Digital Cameras 

Numerous advanced camera writes are utilized as a part of the writing of fingerprint 

frameworks. We arrange them into webcams, cell phone cameras and progressed advanced 

cameras .Webcams are computerized cameras that are generally low in determination and 

exceptionally shoddy in cost. These cameras comes generally implicit work force PCs for 

video visiting purposes. Since these cameras have low determination, it is difficult to get the 

fingerprint points of interest from a picture. Accordingly, numerous examines proposed 

numerous picture pre-handling strategies to defeat this issue. Over the most recent couple of 

years, cell phones cameras have seen an enormous change which have empowered them to be 

utilized to procure fingerprint. Current determination of cell phone have achieved 8 uber 

pixels all things considered. Nonetheless, there are an arrangement of criteria that ought to be 

considered when taking fingerprint pictures utilizing these gadgets, ffor example, lighting, 

remove between the finger and the camera, the proper concentration that ought to be picked, 

and the picture preprocessing steps. Progressed computerized cameras, additionally, have 

been considered and used to secure fingerprint pictures. These cameras delivers high 

determination pictures. In these cameras, picture obtaining is profoundly flexible. They can 

likewise apply propelled picture upgrade and pressure calculations 

FINGERPRINT DATABASES 

In this segment, we portray briefly the absolute most basic fingerprint databases: 

FVC2000 database is utilized as a part of the First International Finger-print Verification 

Competition. The database comprises of four sections; DB1, DB2, DB3 and DB4. DB1 and 

DB2 contain pictures of 110 fingers with eight impressions for every finger (880 pictures 

altogether). The pictures of up to four fingers for every member where taken. Fingerprints of 

the fore and the center fingers of the two hands are taken. Half of the fingerprints are for guys. 

The times of the people in these two databases are 20-30 years of age. DB1 was procured 

utilizing a minimal effort optical sensor (Secure Desktop Scanner) while DB2 was gained an 

ease capacitive sensor (TouchChip by ST Microelectron-ics). As indicated by, these two 

databases can't be utilized to assess the precision of the biometric calculations. That is on the 

grounds that no alerts were taken to ensure the nature of the pictures and the plates of the 

sensor were not methodicallly cleaned after each utilization In DB3, the fingerprints are 

obtained from 19 volunteers, of whom 55% are guys. The pictures of six fingers are taken

from every member (thumb, fore, and center of left and right hands). The periods of the 

people are in the vicinity of 5 and 73. DB3 pictures were obtained utilizing an optical sensor 

(DF-90 by Identicator Technology). DB4 was naturally created utilizing engineered age. 

FVC2002 database was utilized as a part of the Second International Fingerprint Verification 

Competition. It comprises of four sections; DB1, DB2, DB3 and DB4. The quantity of 

pictures is the same as in FVC2000. In this database, four distinct kinds of advances were 

utilized to obtain the fingerprints in the four unique databases (I) an optical sensor 

"TouchView II" by Identix, (ii) an optical sensor "FX2000" by Biometrics, (iii) a capacitive 

sensor "100 SC" by Precise Biometrics, and (iv) manufactured fingerprint age. 90 people took 

an interest in this database gathering.In DB1, DB2, and DB3, pictures of various 30 members 

were gathered. Pictures of 4 fingers (the fore and the center fingers of the two hands) of every 

member were gathered. No any additional exertion were taken to ensure the nature of the 

pictures as the perpos of these pictures are simply to be utilized as a part of rivalry. DB4 was 

artificially created. 

CONCLUSIONS AND FUTURE WORK 

In this paper, we have studied the writing for the most widely recognized fingerprint 

procurement gadgets and the diverse accessible fingerprint picture sets. We have seen that the 

most recent pattern in fingerprint obtaining is to utilize distinctive kinds of advanced cameras 

and uncommonly the utilization of cell phone cameras. We have additionally seen the 

accessible fingerprint dataset obtained utilizing computerized cameras are altogether taken 

with work in debasements due to specific lighting, distinctive foundation, low camera quality, 

picture pressure, and some more. At the end of the day, the accessible fingerprint databases 

needs to a flawless fingerprint database that can be utilized for testing the effect of various 

adjustable debasements. As a future work, we intend to assemble a computerized camera 

procured fingerprint set in idealize conditions that can be redone effectively and can be 

utilized for testing the effect of various debasements on the precision of various fingerprint 

acknowledgment calculations.

A STUDY ON DIGITAL IMAGE PROCESSING 

R.Kowshika & R.Nihitha 


INTRODUCTION 

Image processing is a method to perform some operations on an image, in 

order to get an enhanced image or to extract some useful information from it. It is a 

type of signal processing in which input is an image and output may be image or 

characteristics features associated with that image. Now a days, image processing is 

among rapidly growing technologies. There are two types of methods used for image 

processing namely analogue and digital image processing. Analogue image processing 

can be used for the hard copies like printouts and photographs. Image analysts use 

various fundamentals of interpretation while using these visual techniques. The three 

general phases that all types of data have to undergo while using digital technique are preprocessing 

, enhancement, and display, information extraction. Analog image processing is 

done on analog signals. In this type of processing the images. Digital image processing 

deals with developing a digital system that performs operations on an digital image. 

APPLICATION 

Capturing three-dimensional models has become more and more important during 

last year, due to rapid development of 3D technologies, especially 3Dprinters. There 

are several possibilities of capturing such 3D models ,but the most frequent method is 

3D scanning, which means direct capturing of 3D models. This technology is spreading 

into the new domains and more and more new applications. This paper deals with 

applications of 3D scanning technologies in medicine, where 3D scanning brings 

significant advantages on Medical Volumetry

STORING AND VISUALIZTION OF 3D MODELS 

Each cube contains probability by object eg. Density in case of MRI or CT. 

All object in scan are then approximated by number of same cubes. On the other 

hand, computation over such model are simple and fast. It is useful for models in low 

resolution , which are frequently updated and rebuild. 

MEDICAL OF 3D SCANNING 

The universal of 3D scanning device suitable for medical purposes and 

usable in following abilities: 

High accuracy- essential parameter to being able to distinguish even tiny 

changes of body caused by muscle strengthening. 

Flexibilty since device should be universal , we shall be capable of scanning 

entire body as same as tiny details. Because 3D scanner must be very flexibile. 

Low operational costs- to allow its everyday, use operation shall be 

inexpensive. 

Simple manipulation- device must be as much as possible automated, not 

distributing the personnel with complex setting before each scanning. 

High speed- the scanning procedure must be very fast. In other cases, the 

personnel would not have time to use it and would prefer estimation. 

No limitations- device should be usable with any patient. There should be no 

limitations according to metal parts , health state etc..... 

Harmless operation-using the device shall not be harmful for personnel in any 

circumstances. 

To fulfil all these requirements, the design of medical 3D scanner shall be as 

follows: 

Data capturing sensor shall be structured light 3D scanner or laser scanner due 

to precision reasons. As a result of this, computation of 3D point position. Sensor 

motion shall be motorized from reason of precise sensor localization, automatic 

movement and keeping in measuring range. The problem with computational 

requirements is not serious since the model is once captured and afterwards modified 

occasionally.

Because such device is not commercially available, we created our own 3D scanner 

meeting the specifications. High accuracy is reached by using precise manipulator 

with accurate laser scanner and high flexibility is caused by programmable scanning 

by replaceability of laser scanner, what provides possibility of scanning both tiny and 

large structures. It is 3D modelling system, useful for many different medical 

applications besides. Such models are visualized to operator who defines regions of 

interest(ROI) is defined by method itself. 

TECHNOLOGY USE 

The technology was recently by Sitemens healthineers. MRI scanners equipped 

compressed sensing technologies operate much more quickly than MRI scanners 

currently in use. The result is the first clinical application using compressed sensing 

technology . “compressed sensing “ technology which was developed scans of the 

beating heart can be completed in as few as 25 seconds while the patient breathes 

freely. In contrast in an MRI scanner equipped with conventional acceleration 

techniques must lie still for four minutes or more and hold their breath for as 

manyas seven to 12 times related procedures. In the future, compressed sensing might 

change the way MRI of the abdomen are performed. Today certain populations like 

excluded from having abdominal MRI due to their inability to perform long, 

consecutive and exhausting ,the amount of data required for an image of excellent 

diagnostic quality may be reduced enabling imaging of the abdomen in one continuous 

run. While still in the early stages research reported today in the advances has made 

significant steps towards a new MRI method with enable to personalise life-saving 

medical treatments and allow real-time imaging to take place in locations such as 

operating theatres and GP practises. 

MRI which works by detecting the magnetism of molecules to create an image is a 

crucial tool in medical diagnostics. However current technology is not very efficient- a 

typical hospital scanner will effectively detect only one molecule in every 200,000, 

making it difficult to see the full picture of what happening in the body. 

The research team based at the university of York, has discovered a way to make 

molecules more magnetic and therefore more visible an alternative method which 

could produce a new generation of low-cost and highly sensitive imaging techniques. 

Professor Simon Duckett from the centre for Hyper polarisation in magnetic resonance 

at the university of York said. “what we think we have the potential to achieve with

MRI what could be compared to improvements in computing power and performance 

over the last 40 years. While they are a vital diagnostic tool, current hospital scanners 

could be compared to the abacus, the recent development of more sensitive scanners 

takes us to Alan Turing’s computer and we are now attempting to create something 

scalable and low-cost that would bring us to the tablet or smartphones. 

The research team has found a way to transfer the “invisible” magnetism -a magnetic 

form. Improved scanners are now being trailed in various countries, but because they 

operate in the same way as regular MRI scanners- using a superconducting magnetthese 

new models. 

MEDICAL TECHNOLOGY EQUIPMENT 

PET SCANNER 

PET stands for position Emission Tomography and is a method of body 

scanning that detects radioactive compounds that have been injected into the body to 

provide information on function rather than structure . PET scanners are relatively new 

to the secondary equipment market. 

MRI SCANNER 

Magnetic Resonance Imaging is an imaging technique used primarily in medical 

settings to produce high-quality images of the inside.MRI produce images that are the 

visual equivalent of a slice of anatomy. MRI user radio frequencies a computer and a 

large magnet that surrounds the patient. 

CT SCANNER 

Computerized Axial Tomography scanners use a fan beam of x-rays and a 

detector system that rotates then display on a computer or transferred to film. 

NUCLEAR MEDICINE 

In nuclear medicine diagnosing techniques very small amount of radioactive 

materials. Information from nuclear medicine studied describes organ function not just 

structure.

ULTRASOUND 

This medical imaging techniques uses high-frequency sound waves and their 

echoes. The main advantage is that certain images can be observed without using 

radiation. 

PACS/PICTURE ARCHIVING AND COMMUNICATION TECHNOLOGIES 

These IT-based products improve the speed and consistency of image 

communication with in the radiology department and throughout an enterprise. 

OPINION 

It provides a summary of scientific knowledge on security scanners for 

passenger screening. Although the does per scan arising from the use of screening for 

security purposes are well below the public does limit this does not remove the 

requirements for justification. 

CONCLUSION 

In order to show the degrees of sophistication principles, which are generally at 

the basis of applications. The prospects in this area follows those of scanning. The 

same scanning modality can produce several types of scanning. From this point of 

view magnetic resonance imaging(MRI) is a very good example because different 

sequences give access to the anatomy. 

REFERENCE 

Digital image processing (3 rd edition):Rafael c .Gonzalez . Richard wood 

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3841891/#!Po=16.0550

A STUDY ON DIGITAL IMAGE PROCESSING IN CINEMATOGRAPHY 

J.Roselin Monica,S.Sivalakshmi 


INTRODUCTION 

Digital image processing is the use of computer algorithms to perform image 

processing on digital images. As a subcategory or field of digital signal processing, digital 

image processing has many advantages over analog image processing. It allows a much 

wider range of algorithms to be applied to the input data and can avoid problems such as the 

build-up of noise and signal distortion during processing. Since images are defined over two 

dimensions (perhaps more) digital image processing may be modeled in the form 

of multidimensional systems.Many of the techniques of digital image processing, or digital picture 

processing as it often was called, were developed in the 1960s at the Jet Propulsion 

Laboratory, Massachusetts Institute of Technology, Bell Laboratories, University of Maryland, and a 

few other research facilities, with application to satellite imagery, wire-photo standards 

conversion, medical imaging, videophone, character recognition, and photograph enhancement. The 

cost of processing was fairly high, however, with the computing equipment of that era. That changed 

in the 1970s, when digital image processing proliferated as cheaper computers and dedicated 

hardware became available. Images then could be processed in real time, for some dedicated 

problems such as television standards conversion. As general-purpose computers became faster, they 

started to take over the role of dedicated hardware for all but the most specialized and computerintensive 

operations. With the fast computers and signal processors available in the 2000s, digital 

image processing has become the most common form of image processing and generally, is used 

because it is not only the most versatile method, but also the cheapest.Digital image processing 

technology for medical applications was inducted into the Space Foundation Space 

Technology Hall of Fame in 1994. 

VFX: 

Visual Effects (abbreviated VFX) is the process by which imagery is created or manipulated 

outside the context of a live action shot in film making.Visual effects involve in the 

integration of live-action footage (special effects) and generated imagery (digital effects) to 

create environments which look realistic, but would be dangerous, expensive, impractical, 

time consuming or impossible to capture on film. 

Visual effects using computer-generated imagery (CGI) have recently become accessible to 

the independent filmmaker with the introduction of affordable and easy-touse 

animation and compositing software.Visual effects primarily divides into two groups of:

1. Special Effects: It covers any visual effects that take place in live action, e.g. on set 

explosions or stunt performances. 

2. Digital Effects (commonly shortened to digital FX or FX): It covers the various 

processes by which imagery is created or manipulated with or from photographic 

assets. Digital Effects often involve the integration of still photography and computergenerated 

imagery (CGI) to create environments which look realistic but would be 

dangerous, costly, or impossible to capture in camera. FX is usually associated with 

the still photography world in contrast to visual effects which is associated with 

motion film production. Digital FX also divides into different subgroups of 

professions such as: 

‣ Matte paintings and stills: digital or traditional paintings or photographs which serve 

as background plates for 3D characters, particle effects, digital sets, backgrounds. 

‣ Motion Capture (Mo-Cap for short): It’s the process of recording the movements of 

objects and or people. In a session of motion capture, the subject whose motion is 

being captured is recorded and sampled many times per second by different scanners 

placed all over the environment. There are different types of systems that read the 

actor’s movement. One of which is the optical method that uses tracking cameras that 

lock onto specialized markers placed over the actor’s motion capture suit. The other 

type of method is called the non-optical method where instead of capturing the 

markers location in space, it recorders and measures the inertia and mechanical 

motion in the area. This type of motion capture doesn’t just apply to the body, but can 

be used to track the facial movements and expressions of an actor and transfer them 

to a 3d model later on in the pipeline. The same type of concept of using markers to 

track motion is used, but more often than not, the actor’s face will have painted dots 

on their face rather than ball shaped markers. Not only is the actor’s movements 

recorded in this process, but the movement of the camera is also recorded, which 

allows editors to use this data to enhance the environment the motion captured set is 

imagined in. Once all of this is captured, the motion captured data is mapped to a 

virtual skeleton using software such as Autodesk’s MotionBuilder or other software 

of choice. 

‣ Modelling: Creating 3D models of props or characters using specialised software. 

‣ Animation: Assign movements for any objects and characters in 2D or 3D. 

‣ Compositing: Combining visual elements from different sources to create the illusion 

that all those elements are parts of the same scene.

TECHNOLOGY 

The first animation films can be traced back to the early 1900s which featured 

characters from the popular comic strips of the time. These films made use of the single 

frame method which involved images projected at a high volume of frames per 

second. Gertie the Dinosaur created by sketch artist Winsor McCay in 1914 is believed to be 

the first successful animation short film. 

King Kong, released in 1933 was among the pioneering movies to use this technique. The 

use of miniature models was another Hollywood VFX technique that was in use in the early 

1900s, but it was taken to another level by the classic sci-fi 

franchises StarWars and StarTrek. Hundreds of miniature models manipulated by some 

brilliant camerawork marked these films, which created an unprecedented fan base. 

Superman, released in 1978, was another milestone in the special effects industry. By using 

cables, blue screen and some clever camera tricks, the movie makers created the illusion of a 

flying superhero.Visual effects definitely add value to movies and while Hollywood remains 

the mecca of special effects, film makers the world over are now using VFX to enhance their 

movies. Some of the recent hit films in China to have used VFX heavily include the Zhong 

Kui: Snow Girl and the Dark Crystal, John Woo’s The Crossing, and the recently released 

Monster Hunt, which went on to become the highest grossing Chinese film of all time. In 

India, the larger-than-life Bollywood films use an average of 500 to 1500 VFX shots. Films 

like Dhoom, Chennai Express and the smash-hit Bahubali have used special effects to 

perfection in recent times. The growing popularity of visual effects in world cinema has led 

to a spurt in the demand for special effects companies like Toolbox Studio, which has been 

doing impressive work in the field. 

CONCLUSION 

VFX is a technology which is used for making animated and unbelievable movies 

with its originality. This paper says about the definition, history, and techniques of the VFX. 

This also includes about the lifespan of technology from the starting stage and still now how 

it is useful for flims. VFX also makes a good carrier in life for the persons who have passion 

in cartoon drawing.

REFERENCE 

 

 

 

http://www.toolbox-studio.com/blog/history-of-vfx-in-hollywood/ 

https://en.wikipedia.org/wiki/Digital_image_processing#History 

https://en.wikipedia.org/wiki/Visual_effects

ROAD TRAFFIC SURVEILLANCE AND ACCIDENT FINDING SYSTEM 

BASED ON IOT- A SURVEY 

ABSTRACT 

S.Muralidharan & M.Deepak 

MAM College,Trichy 

Road Traffic is a standout amongst the most essential issue in our creating world. This 

paper presents investigation of various viewpoints and issues identified with the issue. This paper 

underscores on utilizing unmistakable innovation – "Web of Things" (IOT) for creating shrewd 

framework to screen different parameters identified with street movement and utilizing it for 

viable arrangement. 

The overview of the current frameworks and concerned methods identified with the issue zone 

are talked about. Distinctive issues like vehicle identification, impediment recognition, and path 

shot location, mishap discovery and related techniques to comprehend these issues are 

investigated. We propose our "IOT based activity checking and mischance discovery 

framework" comprising of "Raspberry pi", and "pi camera as equipment" which will utilize live 

video as information and process it for social event data about the live movement. The 

framework produces data seeing the activity, for example, number of vehicles, crisis mishap 

circumstances, and dishonorable shot of vehicle. The produced data can be utilized to deal with 

and occupy the live movement as expected to keep away from the issues identified with street 

activity 

Keywords—Internet of Things (IOT); Raspberry pi; Pi camera; Traffic Monitoring;

INTRODUCTION 

Rising road Traffic is one of the most concerning issue handled by the world in 

everyday life. Individuals endure each day in a few or the other route because of 

overwhelming street activity. It is much imperative to consider and enhance the factor of 

street security while contemplating the issue of street activity. The effective answer for this 

issue is to utilize keen computerized innovation for taking care of the continuous movement. 

There has been different research work did to discover the answer for this issue. Yet at the 

same time there is a need of proficient and perfect framework that can be for all intents and 

purposes tried and executed. 

Movement reconnaissance can be extremely helpful for future arranging, for example, 

contamination control, activity flag control, framework and street arranging, inadvertent 

blockage evasion. The information acquired from this framework can be valuable for the 

expectation of the voyaging time and way. Using Internet of Things (IOT) can be 

exceptionally useful for growing such a shrewd framework. IOT is having interconnected 

system of physical parts and gadgets associated for the accumulation, trade and handling of 

different sorts of information for finish of some particular goal. IOT has been very effective 

in expressing in fact brilliant ideas for different spaces associated with this present reality, for 

example, savvy city, shrewd home, keen horticulture, and keen security and so forth. 

This paper exhibits top to bottom portrayal of issues and research issues identified with 

activity observation and mischance identification framework. We additionally exhibit 

examination and discoveries from the investigation of the writing identified with the issue. It 

likewise introduces our proposed framework and future work. The proposed system in 

perspective of IOT is limited with the help of a taking care of board to process the data and a 

camera module to give the live video as information. Raspberry pi board will be used as 

taking care of module and pi camera module will give the data information in video crude 

arrangement h.264 to the Raspberry pi. The framework will distinguish the quantity of 

vehicle cruising by, mischance and anticipate the path shot of the vehicles out and about. The 

foundation subtraction utilizing Gaussian blend model and edge location utilizing watchful 

edge is been executed on Raspberry pi.Motivation 

Significant urban areas of India like Mumbai, Surat and Ahmedabad are confronting the issue 

of street movement gravely. The general population stay stuck in roads turned parking lots 

and gridlocks for a considerable length of time. 

58

There are numerous crisis circumstances on street because of activity which needs moment 

consideration. These crisis riotous circumstances can turn out to be defenseless. In this way it 

is important to direct the activity to guarantee quick stream of the vehicles out and about. 

Subsequently in this developing universe of innovation, the plan to utilize savvy mechanical 

frameworks with wide future extension can be significantly more powerful. 

RESEARCH PROBLEMS 

1. A Research Problems Related to System in Live Environment 

There is wide extent of change for the framework to be executed and tried for all 

intents and purposes. Numerous analysts have proposed different procedures like shrewd 

edge discovery, Spatial-Temporial investigation with Hidden Markov Model (HMM) , 

thresholding and foundation subtraction , 3-D demonstrate and neural system for vehicle 

following, mischance recognition and path shot. They have utilized changing scope of 

equipment like stratix III FPGA, cameras, for example, skillet tilt and zoom camera , 

standard camera, and V8 camcorder at various shifting edge rate and resolutions yet some 

way or another the design of the framework created do not have the accompanying 

components and should be assessed.Ractical compatibility Analysts and technocrats have 

proposed numerous thoughts 

and created numerous mixes of equipment and programming to give most ideal arrangement 

they can.Yet at the same time the frameworks need to give for all intents and purposes 

similarity this present reality for down to earth utilization. A few analysts have worked just 

with disconnected video information which does not give understanding about working of 

their proposed work in genuine world, While some examination work have been tried in 

certifiable which needs more investigation about its working in various clear continuous 

circumstances for down to earth approach with adaptable usage with live movement 

2) Environment and background problem 

This foundation and condition impacts in a few different ways of the gadgets. The 

hindrances out of sight, for example, tree, plants, individuals and so on are factors that can 

influence to the video preparing calculation and consequently can give second rate comes 

about. Edge differential procedure utilized as a part of can have changed surprising 

substantial handling on Raspberry pi board if there is excessively overwhelming movement 

and foundation is intensely thick and if the edge rate isn't set to work with to the irregular 

59

circumstance. Thus the encompassing condition caught in the casing of the camera is a 

critical point to be considered while adopting the pragmatic strategy for this framework. 

3) Software and hardware 

Similarity As depicted before there are numerous products and equipment that has been 

consolidate for a few proposed work. It is extremely imperative for programming to 

commonly work correctly with the equipment. Raspberry pi board isn't intended to work with 

x86 working framework [21] which incorporates windows and a portion of the Linux based 

OS. We liked to utilize Raspbian OS, which is an extraordinarily planned and favored 

working framework for Raspberry pi and python is the advanced dialect that is desirable over 

work with Raspberry pi. The conduct of devices and programming will contrast for various 

equipment, for instance the establishment and working of the Open CV library by Intel as per 

the programming dialect on IOT gadgets for video preparing will vary not at all like in 

ordinary designed PC. Subsequently it is important to discover most ideal mix of the two 

components which will be reasonable to work in live condition with great execution. 

4) Independency 

Most of the current systems which are been observed and tested in the live environment 

needs formation of the design to work as an independent system with remote access and 

control over the network to provide it more compact ability and feasibility to be implemented 

in real world. 

5) Cost effectiveness 

Financially plausible framework will give much better down to earth and achievable 

approach. In the event that we consider greater situation where this sort of framework is 

should have been utilized and kept up on expansive number of streets of the city, cost ends 

up one main consideration. There emerges a need of building framework with minimum cost 

equipment and programming. 

B. Asset limitations for IOT based framework 

1) Processing Power 

The proposed framework will have constrained assets for preparing which will influence 

the handling power. Raspberry pi has just 1 gigabytes of smash and 1.2 gigahertz of Quad 

Core processor. These assets should be overseen and taken care of in an approach to play out 

the different video preparing tasks. Subsequently we will undoubtedly work as indicated by 

this confinement. 

60

2) Power supply 

The propose framework requires consistent power supply yet at current circumstance we 

intend to utilize a battery. So we will have a constrained power supply for actualizing and 

testing our framework. 

3) Atmospheric impact 

The equipment segments of the proposed framework won't have any assurance against 

the climatic circumstances like warmth, rain, and cool and so on which can influence the 

ordinary working of the framework.III. RELATED WORK 

The past work completed by various scientists have been broke down and examined thinking 

about different elements and parameters to produce the vital discoveries of the procedure, 

methods and criteria required for building up the proposed framework. The writing audit has 

been set up into five table speaking to the investigation of systems utilized for vehicle 

discovery, mishap recognition and vehicle following, dataset utilized and sort of the camera 

utilized. The nature of the framework is judged in light of the execution parameter alongside 

the physical equipment that has been utilized for improvement of the framework. The tables 

are readied in light of criteria whether crafted by writers has been executed in live condition 

or tried in disconnected mode. 

A. Vehicle Detection 

The strategies Spatio-transient investigation, Kalman separating Frame distinction , 

Gaussian blend show [20] have been utilized for getting the extraction of the state of vehicle. 

The Gaussian blend demonstrates has been more exact to isolate out the state of moving 

vehicle and it additionally works for versatile foundation. Spatio-transient distinction and 

Kalman separating strategies require complex preparing. Paired Frame contrast is less 

unpredictable yet gives less accuracy contrasted with Gaussian blend demonstrate. 

B. Vehicle Tracking 

Vehicle following gives the path or the way of the moving vehicle. The path of the 

vehicle can be utilized to judge and screen the customary stream of the activity. Vehicle 

following is mostly isolated into two classes’ direction based and optical stream. Optical 

stream based strategy requires low calculation contrasted with direction based. In light of the 

study Lucas-Kanade technique in view of optical stream and K-implies bunching in view of 

direction is discovered more proficient. Relatively Lucas-Kanade needs less calculation and 

gives high precision. The critical examination and discoveries that are acquired The 

investigation about different procedures/strategies was done to discover best methods. The 

61

investigation of the past work helped us to get understanding about the current framework 

and their philosophy. It helped us in seeing how to plan and propose more productive 

framework. 

D. Available Open Source Data Set for Experimentation 

1) VIRAT video informational index 

This informational index contains crude video records of 8.5 hours in .mp4 organize having 

determination of hd 720pixel. The recordings in the informational collection are recorded 

from stationary camera and contain ground data with blend of people strolling and vehicles 

out and about. 

2) MIT Traffic Video Data set 

The MIT movement informational index contains 90 min video succession recorded at 

720x480 utilizing a standard camera. The informational index contains day by day life 

exercises out and about including people. This informational index can consummately be 

utilized for testing and experimentation reason 

PROPOSED SYSTEM 

After the investigation of the related work and examination of the current framework we 

chose to plan a framework which survives and handle a portion of the current research issues 

talked about before. The degree and target of the exploration in light of the elements and 

assets accessible are enrolled underneath. 

A. Extent of Research 

The framework will be made out of raspberry pi and pi camera as equipment gadgets. 

The programming of the framework will be finished utilizing Raspbian OS. The 

programming dialect utilized will be python and openCV library. The framework will work 

with video got in crude arrangement h.264 from raspberry pi camera at 1296 x 730 

determinations 1-49fps or 640 x 480 at 1-90fps. The framework will distinguish 

overwhelming vehicles including trucks, transports, and autos and path of the vehicle will be 

plotted. The vehicle identification will be tested in live condition. The framework will 

distinguish oddity and mischance or uncommon conduct of the vehicle. The mishap 

discovery will be performed in demonstrated/recreated condition or disconnected information 

because of confinement of assets. 

62

B. Goal of Research 

To review break down and think about various procedures identified with vehicle, path 

and mischance discovery for movement reconnaissance. To test the different systems on the 

proposed blend of equipment and check the yield as indicated by various parameters. 

Determination of method for steps portrayed in Frame handling diagram, the overview or 

tried consequences of analyses. To create framework with those chose strategies/calculation 

for identifying number, path and mischance of vehicles with help of information video 

caught from camera at particular required parameter. To take a stab at actualizing and testing, 

this framework in live condition for vehicle location. 

Hardware:- 

1) Raspberry pi 

Raspberry pi is a little measured PC. It has an ARM calculable processor of 1.2 GHZ and 

slam of 1 Gigabyte. It has different working framework as alternative to create different IoT 

ventures. Raspberry pi likewise contains GPIO pins which are additionally programmable for 

different tasks. Because of little size can without much of a stretch be fit for our proposed 

framework. 

2) pi Camera 

Pi camera is a camera module that can be utilized with raspberry pi. There is opening in 

Raspberry pi to straightforwardly associate the pi camera. Pi camera gives input video in 

determination of 640, 720, or 1080 HD. Henceforth it can be effortlessly utilized for our 

motivation. Diverse elements like shine, power can be balanced for Pi camera. 

3) Battery as power source 

4) Internet dongle for remote access 

Programming tools:- 

5) Python 

Python is the authority bolstered dialect for Raspberry pi. There is additionally wide 

network taking a shot at Raspberry and python. The Raspberry pi has been firmly upheld by 

the python programming. Subsequently we work with Python IDE. 

6) Opencv (Intel open source library) 

OpenCV is an open source library for video and picture preparing and it is exceptionally 

productive and powerful for different activities identified with our proposed framework. 

OpenCV is lightweight as far as preparing. 

63

7) Eclipse IDE for remote investigating 

Shroud can be utilized for remote troubleshooting. We can interface remote raspberry and 

Eclipse on neighborhood PC for troubleshooting the projects on raspberry pi. As Eclipse IDE 

has great troubleshooting support consequently it gives us straightforwardness to take a shot 

at Raspberry pi.Architectural diagrams 

Fig . 2. Architecture of Proposed System 

Execute concerned operations on each frames extracted from the live video of pi camera, 

elaborates the processing and operations on single frame. 

D. Experiment and demonstration 

The initial step of our trial was to design the Raspberry pi. The headless establishment of 

the Raspbian OS was done effectively on Raspberry pi. After the establishment of the 

Raspbian OS the establishment of 

the python and openCV was done in the virtual condition to satisfy every one of the 

conditions required. For association of the Raspberry pi with PC/PC remote association 

utilizing secure shell SSH is favored. After the setup of the Raspberry pi we tried three 

strategies in view of our investigation and discoveries for vehicle identification and 

foundation division Frame distinction, Gaussian Mixture Model, Bayesian division. The trial 

64

was done on two dataset VIRAT and MIT Traffic as The strategies were likewise tried on a 

video of Indian street movement with just 360p determination and temperamental camera as 

the most dire outcome imaginable to get the correct data about the working of the three 

methods and discover the best reasonable. We discovered that GMG performed well contrast 

with others and consequently we utilized it with Canny edge identification to check whether 

we can utilize all the more precisely followed edge of moving vehicle for our framework. 

Our future work will incorporate development of bouncing box and impediment end. The 

jumping box is a container shape covering moving vehicle in each edge and it will likewise 

be valuable in distinguishing impediment between two vehicles. We will attempt to explore 

different avenues regarding distinctive techniques for vehicle following and mischance 

location and actualize them to locate the most appropriate one for our objective. We intend to 

test the proposed framework in live condition for vehicle discovery. The framework will be 

tried for remote access over the web utilizing the web dongle and a battery as power source 

giving it an independency. This gets to and controls the framework from any geological area. 

65

CONCLUSION 

A top to bottom review of methodologies utilized by analysts for vehicle discovery, 

vehicle following, and mischance identification is displayed. The accessible open source 

video datasets, helpful for disconnected experimentation, are likewise talked about. From 

study, we reason that (I) Hidden Markov Model and Neural Network strategies are more 

effective and perfect to try different things with mischance discovery as these strategies give 

great exactness, (ii) Lucas-Kanade, a strategy in view of optical stream, and K-implies 

grouping, a system in light of centroid count of the moving article, are helpful for vehicle 

following. The examinations on VIRAT and MIT activity datasets were led and it is reasoned 

that for vehicle identification Gaussian blend model and edge contrast methods demonstrate 

more compelling. The shrewd edge location strategy performed well to find the edge of the 

vehicle. The climate conditions, control supply and preparing power are basic parameters to 

be considered while planning IOT based movement observation framework. In future, we 

intend to create IOT based movement observation framework. We expect to test the 

framework in live condition with a dream of its pragmatic execution in everyday’s life 

REFERENCES 

1. S. Kamijo, Y. Matsushita, K. Ikeuchi and M. Sakauchi. "Traffic monitoring and accident 

detection at intersections." IEEE. Trans. on Intelligent transportation systems, vol. 1, no. 2, 

pp. 108-118, June2000 

2. R. Cucchiara, M. Piccardi, and P. Mello, ” Image Analysis and Rule- Based Reasoning 

for a Traffic Monitoring System” IEEE transactions on Intelligent transportation systems, 

vol. 1, No. 2, pp. 119-130,June2000 

3. C. Lin, J. Tai and K. Song, “Traffic Monitoring Based on Real-Time Image Tracking”, 

IEEE International Conference on Robotics & Automation, Taiwan, pp. 2091-2096, 

September 2003 

4. Y. Jung, K. Lee, and Y. Ho, “Content-Based Event Retrieval Using Semantic Scene 

Interpretation for Automated Traffic Surveillance” IEEEtransactions on Intelligent 

transportation systems, vol. 2, no. 3, pp. 151-163, September 2001 

5. Y. Ki, and D. Lee, “A Traffic Accident Recording and Reporting Model at 

Intersections”,IEEE. Trans. on Intelligent transportationsystems,vol.8, no.2,June2007 

66

STUDY ON IMAGE PROCESSING IN SECURITY USING TOUCH 

SCREEN & VERIFICATION SOFTWARE 

T. Muthu Krithika 

P.Brindha 


ABSTRACT 

Security system with image processing, touch screen and verification software which can 

be used in banks, companies, and at personal secured places.Using Image Processing touch 

screen and verification software more security will be provided than any other systems. This 

system can be used in locker system of banks, companies and at Personal secured places. The 

techniques uses are color processing which use primary filtering to eliminate the unrelated color 

or object in the image. 

INTRODUCTION 

A digital image is a numeric representation, normally binary, of a two-dimensional image. 

Depending on whether the image resolution is fixed, it may be of vector or raster type. By itself, 

the term "digital image" usually refers to raster images or bitmapped images (as opposed to 

vector images).Image processing is often viewed as arbitrarily manipulating an image to achieve 

an aesthetic standard or to support a preferred reality. However, image processing is more 

accurately defined as a means of translation between the human visual system and digital imaging 

devices. 

APPLICATION 

It is very important for banks, companies to provide high security system to their valuable 

items. In this paper by using image processing, touch screen and verification software more 

security will be provided than any other systems. This systemcan be used in locker system of 

banks, companies and at personal secured places. The object detection technique uses color 

processing which use primary filtering to eliminate the unrelated color or object in the image. 

Touch screen and verification software can be used as an extra level of security for customers to 

verify their identify. Customers has to undergo for security verification in three steps, In the first 

67

step, Person is identified using touch screen in which he/she has to touch a point in the touch 

screen which should be same as that of the point recorded initially during the account opening in 

bank. 

In the second step, at the time of account opening in bank, an object is given to the 

customer which is different for different customer which may vary in shape, size or colour and 

hence each object has different pixels value. Verification in second step is done using object 

detection technique of image processing, in which object is verified using camera and object 

should be placed in the screen at the same point as that of point in which it wasinitially placed in 

the screen at the time of account opening. 

In the third step, Verification is done using verification software in which person has to 

answer the details asked by the system. If all the three responses of the person matches with the 

response initially stored in the microcontroller during account opening, the locker security system 

of bank opens otherwise it remains in the closed. This system is more secure than other systems 

because three steps are required for verification. 

TECHNOLOGY 

TOUCH SCREEN POINT DETECTION 

In security system touch screen is used to provide extra security tousers. A touch screen is 

divided into nine points in which atthe time of account opening user will be asked to choose 

anyone point from the eight points, that one point chosen by theuser will be stored in the 

microcontroller 

. At the time of lockeropening in bank, the user will be asked in the first step totouch the 

screen, if he/she touches the correct point of thetouch screen which was initially selected by the 

user at thetime of account opening the system will proceed for the nextstep for object verification 

otherwise the system will not gofor next step, the locker doors will remain closed. After thetouch 

screen point detection, user has to undergo objectdetection which is the second step of this 

system. The ninepoint Touch screen is shown in figure. 

68

OBJECT RECOGNITION 

Object recognition is a process which identifies a specific object in a digital image or 

video. Object recognition algorithms rely on matching, learning, or pattern recognition 

algorithm using appearance-based or feature based techniques. Common techniques include 

edges, gradients, Histogram of Oriented Gradients, Haar wavelets, and linear 

binary patterns. The techniques using are such as color processing which are used as primary 

filtering to eliminate the unrelated color or object in the image. Besides that, shape detection has 

been used where it will use the edge detection, Circular Hough Transform .Identification of 

person is done by matching the data which is initially stored in the microcontroller. Person has to 

verify the object by placing it at the same place on the screen which was initially placed during 

the account opening. If the position and object is verified the system proceeds for the next step. 

69

VERIFICATION SOFTWARE 

Verification software can be used as an extra level of security or customers to verify their 

identify. . At the time of registration of the customer they will be asked to fill details for which 

they have to fill the details that will be stored in themicro controller. All the records of details of 

different customers will be stored in microcontroller that will be done by creating appropriate 

verification software using C or C++.At the time of entering in locker system, different customers 

based on their Touch screen point detection and object recognition and previous information 

stored in micro processor will be asked to enter the verification details, if they will enter all the 

verification details correctly than the locker gate gets open. The customer will be asked to enter 

the password(which will be given initially by the bank) and if the type password is matched with 

password given by bank then the customer will be asked to enter object number (which will be 

given initially at the time of account opening) and if the type object number is matched with 

object number given by bank then the customer will be asked to enter answer of a security 

question (the security question will be chosen by the user at the time of account opening and its 

answer is entered and is stored in the microcontroller) if the given answer matches with the stored 

answer the locker gate gets open otherwise it main in the closed position. 

WORKING PRINCIPLE 

This security system consists of microcontroller, Object detection algorithm, Verification 

Software, keyboard, LED and LCD. The system works on three steps. In the first step, Person is 

Identified using touch screen in which he/she has to touch a point in the touch screen which 

should be same as that of the point recorded initially during the account opening in bank. In 

second step, at the time of account opening in bank, an object is given to the customer who is 

different for different customer who may vary in shape, size or color and hence each object has 

different pixels value. Verification in second step is done using object detection technique of 

image processing, in which object is verified using camera and object should be placed in the 

screen at the same point as that of point in which it was initially placed in the screen at the time of 

account opening. In the third step, Verification is done using verification software in which 

person has to answer the details asked by the system. 

If all the three responses of the person matches with the response initially stored in 

themicro controller during account opening, the locker security system of bank opens otherwise it 

remains in the closed Position. 

70

CONCLUSION 

A banking locker security system using object detection technique, touch screen point 

detection and verification software is implemented. It is more secured system which is cost 

effective. The microcontroller compares with the datastored at the time of account opening. It 

compares the whether the object is same and at the same place as that of initially recorded. It also 

compares the point of touch screen touched with the point selected initially at the time of account 

opening and finally this system verifies details using verification software. 

REFERENCES 

1. Islam, N.S. Wasi-ur-Rahman, M. “An intelligent SMS- based remote Water Metering System”. 

12th International Conference on Computers and Information Technology, 2009, 21-23 Dec. 

2009, Dhaka, Bangladesh. 

2. International Conference on Robotics, Vision, Information and Signal Processing 2007 

(ROVISP2007), Penang, 28 – 30 November 2007. 

71

A STUDY ON DIGITAL CAMERA IMAGE 

FORENSIC METHODS 

K.Muthulakshmi & S.Anjala 

Fatima College,Madurai 

ABSTRACT 

The aim of this survey is to provide a comprehensive overview of the state of the 

art in the area of image forensics. These techniques have been designed to identify the source of 

a digital image or to determine whether the content is authentic or modified, without the 

knowledge of any prior information about the image under analysis. All the tools work by 

detecting the presence, the absence, or the incongruence of some traces intrinsically tied to the 

digital image by the acquisition device and by any other operation after its creation. The paper 

has been organized by classifying the tools according to the position in the history of the digital 

image in which the relative footprint is left: acquisition based methods, coding –based methods, 

and editing –based schemes. 

There are two main interest in digital camera image forensics, namely source 

identification and forgery detection. In this paper, we first briefly provide an introduction to the 

major processing stages inside a digital camera and the review several methods for source 

Digital camera identification and forgery detection. Existing methods for source identification 

explore the various processing Stages inside a digital camera to device the clues for 

distinguishing the source cameras while forgery detection checks for Consistencies in image 

quality are for present for contain characteristics as evidence of tampering. 

The growth sophisticated image processing and editing software has made the 

manipulation of digital images easy and imperceptible to the naked eyes. This increased the 

demand to assess the trustworthiness of digital images when used in crime investigation, as 

evidence in court of law and for surveillance purposes. This paper presents a comprehensive 

investigation of the progress and challenges in the field of digital image forensics to help the 

beginners in developing understanding, apprehending the requirements and identifying the 

research gaps in this domain. 

72

INTRODUCTION 

In digital image processing, computer algorithms are used to perform image processing. 

Digital image processing has several advantages over the analog image processing. It provides a 

large number of algorithms to be used with the input data. In digital image processing we can 

avoid some processing problems such as noise creation and signal distortion at some point in 

signal processing. In 2000s, fast computers became available for signal processing and digital 

image processing has become the popular form of image processing. For the reason that, signal 

image processing became versatile method, and also cheapest. 

The term digital image processing generally refers to processing of a two-dimensional picture 

by a digital computer. In a broader context, it implies digital processing of any two-dimensional 

data. A digital image is an array of real numbers represented by a finite number of bits. The 

principle advantage of Digital Image Processing methods is its versatility, repeatability and the 

preservation of original data precision. 

In digital image processing, we are concentrating on one specific application. That is digital 

camera image forensic methods. 

Multimedia Forensics has become important in the last few years. There are two main interests, 

namely source identification and forgery detection. Source identification focuses on identifying 

the source digital devices (cameras, mobile phones, camcorders, etc) using the media produced 

by them, while forgery detection attempts to discover evidence of tampering by assessing the 

authenticity of the digital media (audio clips, video clips, images, etc). 

A digital camera or digicam is a camera that captures photographs in digital memory. Most 

cameras produced today are digital, [1] and while there are still compact cameras on the 

market, the use of dedicated digital cameras is dwindling, as digital cameras are now 

incorporated into many devices ranging from mobile devices to vehicles. However, high-end, 

high-definition dedicated cameras are still commonly used by professionals. 

73

TECHNOLOGIES USED IN DIGITAL CAMERA 

1. ELEMENTS OF A TYPICAL DIGITAL CAMERA 

Digital cameras consist of lens system, filters, and color filter array (CFA), image sensor, and 

digital image processor (DIP). Color images may suffer from aberrations caused by the 

lenses, such as, 

 

 

Chromatic aberration and 

Spherical aberration. 

2. CAPTURING ON IMAGE 

In capturing on image, First, We have to capture on 3d image using a digital camera and send 

it to a particular system (digital image processing system) to focus on some specific area, and 

then it will produce an output as a zoomed image. Here, it is focused on a water drop in a leaf. 

74

3. TYPES OF DIGITAL CAMERA 

 

 

 

 

 

Compact digital camera 

Bridge camera 

Mirror less interchangeable-lens camera 

Digital single-lens reflex(DSLR) camera 

Digital single-lens translucent(DSLT) camera 

4. DIGITAL IMAGE FORENSIC METHODS 

There are some techniques that use to detect the digital image forensic methods. They 

are, 

1. Using JPEG Quantization Tables 

2. Using Chromatic Aberration 

3. Using Lighting 

4. Using Camera Response Function (CRF) 

5. Using Bicoherence and Higher Order Statistics 

6. Using Robust Matching 

5. FLOWCHART--DIGITAL IMAGE FORENSIC METHODS 

75

The forensic analysis of digital images has become more significant in determining the 

origin and authenticity of a photograph.The trustworthiness of photographs has an essential role 

in many areas, including: forensic investigation, criminal investigation, surveillance systems, 

intelligence services, medical imaging, and journalism.Digital images can be captured by some 

digital cameras or scanners and can be generated on computers too. Passive image forensic 

techniques for source identification work on the basic assumption that the fingerprints of the 

imaging sensors, in-camera processing operations and compression are always present in images. 

Detection of camera specific fingerprints identifies the image capturing device and justify that the 

image is not computer rendered. The two images having the same in-camera fingerprints are 

judged to be taken by the same device. The absence of fingerprints in images suggests that either 

the image is computer generated or has been maliciously tampered thereby calling for image 

integrity verification. Based on the above assumptions the published works are presented in this 

section with respect to two issues: firstly, to distinguish between the natural and computer 

generated images; and secondly, to identify the image capturing device if the image is natural. 

Image tampering is a deliberate attempt to add, remove or hide some important details of 

an image without leaving any obvious traces of the manipulation. The digital images are 

generally tampered by region duplication, image splicing or image retouching. Region 

duplication is also recognized as cloning or copy-move attack, where selective regions from an 

image are copied, sometimes transformed, and then pasted to new locations within the image 

itself with the main aim of concealing some original image contents. Image splicing on the other 

hand uses selected regions from two or more images to be pasted together for producing a new 

image. 

CONCLUSION 

With the outgrowth of the imaging and communication technology, the exchange of 

digital images has become easy and extensive. But at the same time, the instances of 

manipulations in the digital images have also increased thereby resulting in greater need for 

establishing ownership and authentication of the media. Digital image forensic researcher 

community is continuously attempting to develop techniques for detection of the imaging device 

used for image acquisition, tracing the processing history of the digital image and locating the 

region of tampering in the digital images. The sensor, operational and compression fingerprints 

have been studied with various image features to achieve the purposes. An attempt to recover the 

tampered region details is expected to be an appealing investigation domain for many researchers. 

76

Most of the work done in image forensics has focused on detecting the fingerprints of a specific 

kind of tampering operation. But, practically a manipulated image is often the result of multiple 

such tampering operations applied together. Thus, the need is to develop a technique or 

framework capable of detecting multiple attacks and tampering. 

REFERENCES 

 

 

 

www.researchgate.net/publication/299367087_Digital_Image_Forensics_Progress_and_C 

hallenges 

http://www.comp.nus.edu.sg/~mohan/papers/forensic_surv.pdf 

 

Z. J. Geradts, J. Bijhold, M. Kieft, K. Kurosawa, K. Kuroki, and N. Saitoh, “Methods for 

Identification of Images Acquired with Digital Cameras”, Proc. of SPIE, Enabling 

Technologies for Law Enforcement and Security, vol. 4232, February 2001. 

 

J. Lukas, J. Fridrich, and M. Goljan, “Digital Camera Identification from Sensor Pattern 

Noise”, IEEE Transactions on Information Forensics and Security, June 2006. 

 

Cheddad A. Doctored Image Detection: A Brief Introduction to Digital Image Forensics. 

Inspire magazine, July 2012. 

77

A STUDY ON RAINFALL PREDICTION 

S.Alamelu & S.Nasreen Farzhana 

Fatima College(Autonomuos),Madurai 

ABSTRACT 

Water is one of the most important substances on earth. All plants and animals must have 

water to survive. If there was no water there would be no life on earth.This paper represents an 

analysis about rainfall prediction using Digital image processing. The proposed approach here is 

to use the digital cloud images to predict rainfall. Considering the cost factors and security issues, 

it is better to predict rainfall from digital cloud images rather than satellite images. The status of 

sky is found using wavelet. The status of cloud is found using the Cloud Mask Algorithm. The 

type of cloud can be evolved using the K-Means Clustering technique. The type of rainfall cloud 

is predicted by analyzing the color and density of the cloud images. The result predicts the type of 

cloud with its information like classification, appearance and altitude and will provide the status 

of the rainfall. 

INTRODUCTION 

Digital image processing is the processing of digital images with computer algorithms.A 

digital image is nothing more than a two dimensional signal. It is defined by the mathematical 

function f(x,y) where x and y are the two co-ordinates horizontally and vertically. The value of 

f(x,y) at any point is given as the pixel value at that point of an image. 

Digital image processing is the use of computer algorithms to create, process, communicate, and 

display digital images. Digital image processing algorithms can be used to convert signals from 

an image sensor into digital images, to improve clarity and remove noise and other artifacts, to 

extract the size, scale, or number of objects in a scene, to prepare images for display or printing 

and to compress images for communication across a network. 

A digital image is formed as follows. Since capturing an image from a camera is a physical 

process. The sunlight is used as a source of energy. A sensor array is used for the acquisition of 

the image. So when the sunlight falls upon the object, then the amount of light reflected by that 

object is sensed by the sensors, and a continuous voltage signal is generated by the amount of 

sensed data. In order to create a digital image, we need to convert this data into a digital form. 

This involves sampling and quantization. Sampling and quantization result in a two dimensional 

array or matrix of numbers which are nothing but a digital image. 

78

RAINFALL PREDICTION 

The paper is about “Rainfall Prediction Using Digital Image Processing”. The main 

aim is to use digital cloud image for estimating rainfall and to detect the type of clouds using 

different image processing techniques. Water is a life giver - even a life creator. A life without 

water is unimaginable. Every single cell in the body requires water in order to function properly. 

One important way in which the bodily tissues use water, for instance, is to regulate the body 

temperature. Hydraulic power is a wonderful source of green energy, where power is created out 

of water. Water is used in agriculture for irrigating fields of crops. A large amount of fresh water 

is got from rain. Rainfall is one such important component for life. Rainfall play vital role in our 

country’s social and economic development. 

Rainfall being that much important, how are they formed as follows. Rain that’s fallen 

may have been water in the ocean a couple of days before. The water that is at the top of oceans, 

rivers and lakes turns into water vapor in the atmosphere using energy from the sun. The water 

vapor rises in the atmosphere and there it cools down and forms tiny water droplets through 

something called condensation. These then turn into clouds. When they all combine together, 

they grow bigger and are too heavy to stay up there in the air. This is when they will fall to the 

ground as rain. 

Rainfall can also be predicted with the use of satellite, but it the costliest method. So 

digitalimage processing techniques are used to determine rainfall. With a prior prediction of 

rainfall, the people can find some way to save the water for future usage. Accurately forecasting 

heavy rainfall can allow for warning of floods before rainfall occurs. Additionally, such 

information is useful in agriculture to improve irrigation practices and the effectiveness of 

applying fertilizer, pesticides, and herbicides to crops. Digitalimage processing is one such useful 

technique to predict rainfall. 

TECHNOLOGY USED FOR THE RAINFALL PREDICTION 

Predicting the rainfall consists of six phases. In the first phase data is collected. In the 

second phase the status of sky is found. In the third phase the status of cloud is found. In the 

fourth phase the type of cloud is evolved. In the fifth phase the information about the cloud and 

the status of rain are displayed. In the sixth phase the analysis and measurement takes place. 

79

Data 

collection 

Sky status 

Cloud 

status 

Cloud type 

Rainfall 

estimation 

I. Data collection: The data collected is the digital cloud images. A digital camera is used to 

capture the clouds. The image is stored in the file system. The dimension chosen for the images is 

400 x 250. The format used is *.jpeg. 

II. Sky status: The second step is sky status. The wavelet is used for finding the sky status. A 

wavelet is nothing but a small wave of water. It separates the point needed for the cluster. In 

wavelet separation points are used in the identification of the clouds. The wavelet threshold for 

the clouds is > 50 and

Also the cloud mask algorithm is used here. The cloud mask algorithm consists of certain tests. 

Single pixel threshold tests are used first. Dynamic histogram analysis is used to get threshold. 

Thick High Clouds (Group 1): Thick high clouds are detected with threshold tests that rely on 

brightness temperature in water vapor bands and infrared. Thin Clouds (Group 2): Thin clouds 

tests rely on brightness temperature difference tests. Low Clouds (Group 3): Low clouds are best 

detected using solar reflectance test and brightness temperature difference. Spatial uniformity test 

is also used over land surface. High Thin Clouds (Group 4): This test is similar to Group1, but it 

is spectrally turned to detect the presence of thin cirrus. Spatial uniformity test and Brightness 

temperature difference test are applied. Temporal uniformity test is also processed. 

IV. 

Cloud type: The major task is to find the type of cloud as per the cloud status. Each and every 

cloud will be having its own shape and density and the values are matched accordingly. The type 

of cloud is identified by using clustering. We use K-means clustering to combine the pixels in 

order to differentiate the clouds. The thickness of the clouds will be in the base part. The color, 

Shape and Texture are the concepts used in order to find the type of cloud. The formula to find 

the cloud type is shown as follows: 

H(n) = ∑ C[i,j] , Cloud id = Highest Density of Cloud Status 

K-Means algorithm is an unsupervised clustering algorithm that classifies the input data points 

into multiple classes based on their inherent distance from each other. The algorithm assumes that 

the d ata features form a vector space and tries to find natural clustering in them. The points are 

clustered around centroids μi∀=1….k which are obtained by minimizing the objective. An 

iterative version of the algorithm can be implemented. The algorithm takes a 2 dimensional image 

as input. Various steps in the algorithm are as follows: 

Compute the intensity distribution (also called the histogram) of the intensities. 

The Centroids are initializing with the k random intensities. 

Repeat the below steps until the cluster a label of the image does not change anymore. 

Cluster the points based on distance of their intensities from the centroid intensities. 

Cluster the points based on distance of their intensities from the centroid intensities. 

Compute the new centroid for each of the clusters. 

V. Rainfall estimation: The major step is the estimation of rainfall is estimated as per the type we 

recognize. There are different types of clouds. They are as follows: 

81

High Clouds: 

Cirrus - The ice-crystal cloud is a feathery white cloud that is the highest in the sky. It has a 

wispy looking tail that streaks across the sky and is called a fall streak. 

Cirrostratus - A milky sheet of ice-crystal cloud that spreads across the sky. 

Low clouds: 

Cumulus - The fair weather cloud. Each cloud looks like a cauliflower. 

Stratocumulus - A layer of cloud that can sometimes block the sun. 

Rain Clouds: 

Cumulonimbus and Nimbostratus - The dark, rain carrying clouds of bad weather. They are to be 

blamed for most of the winter rains and some of the summer ones. They cover the sky and block 

the sun. 

Cumulonimbus and nimbostratus are the rainfall clouds. So we take the color and shape and also 

the width and find the rainfall status. The temperature is also taken into account. Cloud 

information gives the theoretical proof of cloud that is altitude, height, appearance and 

classification are given. 

CONCLUSION 

The rainfall would be estimate accurately by determining type of cloud using the methods 

like pca or contourlet, cloud screening algorithm and k-mean clustering. Theoretical study says 

that it must provide better performance compared to other techniques like wavelet and other. 

Considering the cost factors and security issues, the digital cloud images were used to predict 

rainfall rather than satellite images. Prediction of rainfall can also be done with the help of neural 

networks and artificial intelligence. 

82

BIBLIOGRAPHY 

 

 

 

 

 

 

 

 

 

 

Watch "Introduction to Digital Image Processing by Ms. Geetanjali Raj [Digital Image 

Processing]" on YouTube 

https://youtu.be/NZpAb6_Yzbg 

Watch "Types of Clouds - Cirrus, Cumulus, Stratus, Nimbus | UPSC IAS Geography" on 

YouTube 

https://youtu.be/BC6t_QOtVu8 

Watch "Technical Course: Cluster Analysis: K-Means Algorithm for Clustering" on 

YouTube 

https://youtu.be/aiJ8II94qck 

Prediction of rainfall using image processing 

https://www.slideshare.net/mobile/VineeshKumar1/prediction-of-rainfall-using-imageprocessing 

Rain Fall Using Cloud Images | Cluster Analysis | Cloud 

https://www.scribd.com/document/146111231/Rain-Fall-Using-Cloud-Im 

83

A STUDY ON PLANT DISEASE IDENTIFICATION 

A.Gnana Priya & B.Muthulakshmi 

Fatima College(Autonomuos),Madurai 

ABSTRACT: 

India is an agricultural country. More than 70 percent of the people depend on agriculture. 

Growing up the crop and production is a challengeable task. Because, many of the crops are 

affected by pest. Insecticide is one of the best remedy for pest attack. But it is sometimes 

dangerous for birds, animals and also for humans. Some crops requires close monitoring that 

helps in management of diseases. Now a day’s digital image processing is widely used in 

agricultural field. This method helps to identify the parts of the plant and find out the disease or 

pest as earlier. It is also useful for understanding the better relationship between the climatic 

condition and disease. 

Image processing techniques could be applied on various applications as follows: 

1. To detect plant leaf, stem, and fruit diseases. 2. To quantify affected area by disease. 3. 

To find the boundaries of the affected area. 4. To determine the color of the affected area 5.To 

determine size & shape of fruit.A counties wealth is detected by agriculture. Disease in plants 

causes major loss in an economy. DIP is the use of computer algorithms to create process, 

communicate, and display digital images. We may conclude that DIP is a useful and effective 

technique for the crop cultivation. 

Disease is caused by pathogen which is any agent causing disease. In most of the cases 

pests or diseases are seen on the leaves or stems of the plant. Therefore identification of plants, 

leaves, stems and finding out the pest or diseases, percentage of the pest or disease incidence, 

symptoms of the pest or disease attack, plays a key role in successful cultivation of crops. 

84

INTRODUCTION 

In biological science, sometimes thousands of images are generated in a single 

experiment. These images can be required for further studies like classifying lesion, scoring 

quantitative traits, calculating area eaten by insects, etc. Almost all of these tasks are processed 

manually or with distinct software packages two major issue excessive processing time and 

subjectiveness rising from different individuals. Hence to conduct high throughput experiments, 

plant biologist need efficient computer software to automatically extract and analyze significant 

content. 

Machine learning-based detection and recognition of plant diseases can provide extensive 

clues to identify and treat the diseases in its very early stages. Comparatively, visually or naked 

eye identification of plant diseases is quite expensive, inefficient, inaccurate and difficult. 

Automatic detection of plant diseases is very important to research topic as it may prove the 

benefits in monitoring large fields of crops, and thus automatically detect the symptoms of 

diseases as soon as they appear on plant leaves.Initially, the infected Plant leaf images are taken 

as input image and image is passed to pre-processing step, to resize the image and remove the 

noise content by using a median filter. At the next stage, the pre-processed image is passed to the 

segmentation for partitioning into clusters. FCM clustering technique is used for segmentation 

which is very fast, flexible and easy to implement than others. Further, the segmented image 

extracts the features of an image by using methods like color correlogram, SGLDM, and Otsu 

methods. Finally, the classifier is used for classification and recognition of plant disease. One of 

the best classifiers is SVM, which is more accurate than others and then it is stored in the 

knowledge base 

85

PROPOSEDMETHODOLOGY: 

1.Image Acquisition: First we need to select the plant which is affected by the disease and then 

collect the leaf of the plant and take a snapshot of leaf and load the leaf image into the system. 

2.Segmentation: It meansrepresentation of the image in more meaningful and easy to analyse 

way. In segmentation a digital image is partitioned into multiple segments can defined as superpixels. 

Low Contrast: image pixel values are concentrated near a narrow range. 

Contrast Enhancement: In figure.2, the original image is the image given to the system and the 

output of the system after contrast enhancement. 

3. Feature extractions, is the process done after segmentation. According to the segmented 

information and predefined dataset some features of the image should be extracted. This 

extraction could be the any of statistical, structural, fractal or signal processing. Color cooccurrence 

Method, Grey Level Co-occurrence Matrices (GLCM), Spatial Gray-level 

Dependence Matrices (SGDM) method, Gabor Filters, Wavelets Transform and Principal 

component analysis are some methods used for feature extraction 

4. Classification of diseases 

Classification technique is used for training and testing to detect the type of leaf disease. 

Classification deals with associating a given input with one of the distinct class. In the given 

system support vector machine [SVM] is used for classification of leaf disease. The classification 

process is useful for early detection of disease, identifying the nutrient deficiency. 

Based on classification leaves are mainly affected with fungal, bacterial and viral. The 

following describes common symptoms of fungal, bacterial and viral plant leaf diseases. 

a) Bacterial disease symptoms: 

The disease is characterized by yellowish green spots which come into view as watersoaked. 

The lesions amass and then appear as dry dead spot 

86

) Viral disease symptoms: 

Among all plant leaf diseases, those caused by viruses are most complicated to diagnose. 

All viral disease presents some degree of reduction in virus-infected plants. The production 

length of such infected is usually short. This virus looks yellow or green stripes or spots on 

foliage. Leaves might be wrinkled, curled and growth may be stunned as depicte 

c) Fungal disease symptoms: 

It is a type of plant pathogen, they are responsible for serious plant diseases. They damage 

plants by killing cells. It disseminates through the wind, water, and movements of contaminated 

soil, animals, birds etc. In the initial stage, it appears on older leaves as water-soaked, gray-green 

spots. Later these spots darken and then white fungal growth forms on the undersides 

CONCLUSION 

We would like to conclude that this is an efficient and accurate technique for a detection 

of plant diseases. In our research, plant disease is detected by SVM classifiers. The SVM 

classifiers are based on color, texture and shape features. The algorithm in the proposed approach 

uses image segmentation and classification technique for the detection of plant disease. The 

accurate Disease detection and classification of the plant leaf image is very important for the 

successful cultivation of cropping and this can be done using image processing. This paper 

discussed various techniques to segment the disease part of the plant. Hence there is working on 

development of fast, automatic, efficient and accurate system, which is use for detection disease 

on unhealthy leaf. Also Comparison of different techniques of digital image processing is done 

which gives the different results on different databases. Work can be extended for development of 

system which identifies various pests and leaf diseases also. 

87

REFERENCES 

1.)SujeetVarshneyet al, International Journal of Computer Science and Mobile Computing, 

Vol.5 Issue.5, May- 2016, pg. 394-398 

2.) International Journal of Computational Intelligence Research ISSN 0973-1873 Volume 13, 

Number 7 (2017), pp. 1821-1828 

3.) Leaf disease detection using image processing Sujatha R*, Y Sravan Kumar and Garine 

Uma Akhil School of Information Technology and Engineering, VIT University, Vellore 

4.) Surender Kumar Department of CSE Chandigarh University Gharuan, Punjab 

International Journal of Computer Applications (0975 – 8887) 

5.) S. Megha, R C. Niveditha, N. SowmyaShree, K. Vidhya ; International Journal of Advance 

Research, Ideas and Innovations in Technology. 

6.) Journal of Advanced Bioinformatics Applications and Research ISSN 0976-2604 Vol 2, Issue 

2, June-2011, pp 135-14 

88

ANALYSIS ON NETWORK SECURITY 

S.Karthick Raja & S.Praveenkumar 

NMSSVN College,Madurai 

ABSTRACT: 

In the past decades, computer networks were primarily used by the researches for sending 

e-mails and by corporate employees for sharing the printers. While using network systems for 

theses utilities, security was not a major threat and did not get the due attention .In today’s 

world computer networks gained an immensely and cover a multitude of sins. It covers simple 

issues like sending hate mails. Security problems also are very severe like stealing the 

research papers of the recent discoveries and inventions by the scientists, who use internet as 

a sharing tool, also hacking the financial products like credit cards, debit cards, bank accounts 

etc by hacking the passwords and misusing the accounts. Cryptography is the ancient science 

of encoding messages so that only the sender and receiver can understand them. 

Cryptography can perform more mathematical operations in a second than a human being 

could do in a lifetime. There are three types of cryptographic schemes. They are: 

Secret Key Cryptography (SKC) 

Public Key Cryptography (PKC) 

Hash Functions 

INTRODUCTION 

A basic understanding of computer networks is requisite in order to understand the 

principles of network security. The Internet is a valuable resource, and connection to it is 

essential for business, industry, and education. Building a network that will connect to the 

Internet requires careful planning. Even for the individual user some planning and decisions 

are necessary. The computer itself must be considered, as well as the device itself that makes 

the connection to the local-area network (LAN), such as the network interface card or modem. 

The correct protocol must be configured so that the computer can connect to the Internet. 

Proper selection of a web browser is also important 

89

What Is A Network? 

A “network” can be defined as “any set of interlinking lines resembling a net, a 

network or roads||an interconnected system, a network is simply a system of interconnected 

computers and how they are connected is irrelevant. 

The International Standards Organization (ISO) Open System Interconnect (OSI) model 

defines internetworking in terms of a vertical stack of seven layers. The upper layers of the 

OSI model represent software that implements network services like encryption and 

connection management. The lower layers of the OSI model 

implement more primitive, hardware-oriented functions like routing, addressing, and flow 

control. 

SECURITY SERVICES 

X.800 defines it as: a service provided by a protocol layer of communicating open systems, 

which ensures adequate security of the systems or of data transfers 

RFC 2828 defines it as: a processing or communication service provided by a system to give 

a specific kind of protection to system resources 

X.800 defines it in 5 major categories 

Authentication - assurance that the communicating entity is the one claimed 

Access Control - prevention of the unauthorized use of a resource 

Data Confidentiality –protection of data from unauthorized disclosure 

90

Data Integrity - assurance that data received is as sent by an authorized entity 

Non-Repudiation - protection against denial by one of the parties in a communication 

SECURITY ATTACKS 

Passive attacks - eavesdropping on, or monitoring of, transmissions to: 

obtain message contents, or monitor traffic flows 

Active attacks – modification of data stream to: 

Masquerade- of one entity as some other replay previous messages modify messages in 

transitdenial of service 

ENCRYPTION 

Symmetric encryption or conventional / private-key / single-key sender and recipient 

share a common key all classical encryption algorithms are 

Private-key was only type prior to invention of public-key in 1970’s 

Symmetric Cipher Model: 

CRYPTOGRAPHY 

Cryptography can be characterized by: 

Type of encryption operations used--substitution / transposition / product 

number of keys used--single-key or private / two-key or public way in which plaintext is 

processed--block / stream 

TYPES OF CRYPTANALYTIC ATTACKS: 

‣ Cipher text only--only know algorithm / cipher text, statistical, can identify plaintext 

‣ known plaintext --know/suspect plaintext &cipher text to attack cipher 

‣ chosen plaintext --select plaintext and obtain cipher text to attack cipher 

‣ chosen cipher text --select cipher text and obtain plaintext to attack cipher 

‣ chosen text --select either plaintext or cipher text to en/decrypt to attack cipher 

91

DES 

The Data Encryption Standard (DES) is a cipher (a method for encrypting 

information) that was selected by NBS as an official Federal Information Processing Standard 

(FIPS) for the United States in 1976 and which has subsequently enjoyed widespread use 

internationally. It is based on a Symmetric-key algorithm that uses a 56-bit key. The 

algorithm was initially controversial with classified design elements, a relatively short key 

length, and suspicions about a National Security Agency (NSA) backdoor. DES consequently 

came under intense academic scrutiny which motivated the modern understanding of block 

ciphers and their cryptanalysis. 

DES is now considered to be insecure for many applications. This is chiefly due to the 56-bit 

key size being too small; in January, 1999, distributed.net and the Electronic Frontier 

Foundation collaborated to publicly break a DES key in 22 hours and 15 minutes (see 

chronology). There are also some analytical results which demonstrate theoretical weaknesses 

in the cipher, although they are unfeasible to mount in practice. The algorithm is believed to 

be practically secure in the form of Triple DES, although there are theoretical attacks. In 

recent years, the cipher has been superseded by theAES 

DES is the archetypal block cipher — an algorithm that takes a fixed-length string of 

plaintext bits and transforms it through a series of complicated operations into another cipher 

text bit string of the same length. In the case of DES, the block size is 64 bits. DES also uses a 

key to customize the transformation, so that decryption can supposedly only be performed by 

92

those who know the particular key used to encrypt. The key ostensibly consists of 64 bits; 

however, only 56 of these are actually used by the algorithm. Eight bits are used solely for 

checking parity, and are thereafter discarded. Hence the effective key length is 56 bits, and it 

is usually quoted as such. 

Like other block ciphers, DES by itself is not a secure means of encryption but must instead 

be used in a mode of operation. FIPS-81 specifies several modes for use with DES. Further 

comments on the usage of DES are contained in FIPS-74. [15] 

The algorithm's overall structure is there are 16 identical stages of processing, termed rounds. 

There is also an initial and final permutation, termed IP and FP, which are inverses (IP 

"undoes" the action of FP, and vice versa). IP and FP have almost no cryptographic 

significance, but were apparently included in order to facilitate loading blocks in and out of 

mid-1970s hardware, as well as to make DES run slower in software. 

Before the main rounds, the block is divided into two 32-bit halves and processed 

alternately; this criss-crossing is known as the Feistel scheme. The Feistel structure ensures 

that decryption and encryption are very similar processes — the only difference is that the 

subkeys are applied in the reverse order when decrypting. The rest of the algorithm is 

identical. This greatly simplifies implementation, particularly in hardware, as there is no need 

for separate encryption and decryption algorithms. 

The symbol denotes the exclusive-OR (XOR) operation. The F-function scrambles 

half a block together with some of the key. The output from the F-function is then combined 

with the other half of the block, and the halves are swapped before the next round. After the 

final round, the halves are not swapped; this is a feature of the Feistel structure which makes 

encryption and decryption similar processes. 

93

The key-schedule of DES 

The key schedule for encryption — the algorithm which generates the subkeys. 

Initially, 56 bits of the key are selected from the initial 64 by Permuted Choice 1 (PC-1) — 

the remaining eight bits are either discarded or used as parity check bits. The 56 bits are then 

divided into two 28-bit halves; each half is thereafter treated separately. In successive rounds, 

both halves are rotated left by one and two bits (specified for each round), and then 48subkey 

bits are selected by Permuted Choice 2 (PC-2) — 24 bits from the left half, and 24 from the 

right. The rotations (denoted by "

CONCLUSION 

It's important to build systems and networks in such a way that the user is not 

constantly reminded of the security system. 

Developers need to evaluate what is needed along with development costs, speed of 

execution, royalty payments, and security strengths. That said, it clearly makes sense to use as 

strong security as possible, consistent with other factors and taking account of the expected 

life of the application. Faster computers mean that longer keys can be processed rapidly but 

also mean that short keys in legacy systems can be more easily broken. 

It's also extremely important to look at the methods of applying particular algorithms, 

recognizing that simple applications may not be very secure. Related to this is the issue of 

allowing public scrutiny, something that is essential in ensuring confidence in the product. 

Any developer or software publisher who resists making the cryptographic elements of their 

application publicly available simply doesn't deserve trust and is almost certainly supplying 

an inferior product. 

Secure communication over insecure channels is the objective of this concept. The claim of 

complete security is substantiated to a large extent. Thus a detailed study of Cryptography & 

Network Security is reflected in this presentation 

95

REFERENCES 

Computer Networking-Tanenbaum. 

Cryptography and Network Security-William Stallings 

Eli Biham: A Fast New DES Implementation in Software Cracking DES: Secrets of 

Encryption Research, Wiretap Politics, and Chip Design, Electronic Frontier 

Foundation 

A.Biryukov, C. De Canniere, M. Quisquater (2004)."On Multiple Linear 

Approximations".Lecture Notes in Computer Science3152: 1–22.doi:10.1007/b99099. 

http://www.springerlink.com/content/16udaqwwl9ffrtxt/. (preprint). 

Keith W. Campbell, Michael J. Wiener: DES is not a Group. CRYPTO 1992: pp512– 

520 

Don Coppersmith. (1994). The data encryption standard (DES) and its strength against 

attacks. IBM Journal of Research and Development, 38(3), 243–250. [1] 

Whitfield Diffie, Martin Hellman, "Exhaustive Cryptanalysis of the NBS Data 

Encryption Standard" IEEE Computer 10(6), June 1977, pp74–84 

WEBSITES 

www.rsasecurity.com 

www.itsecurity.com 

www.cryptographyworld.com 

96

A STUDY ON DIGITAL IMAGE PROCESSING INBRAIN TUMOR 

DETECTION 

INTRODUCTION 

SINDHUJA& 

P. M. SONIA PRIYA 

The aim o digital image processing is to improve the pictorial information for human 

interpretation; and processing image data for storage transmission, and representation for 

autonomous machine perception. It is the use of computer algorithms to perform image 

processing on digital images. It has many advantages over analog image processing. It allows a 

much wider range of algorithms to be applied to the input data and can avoid problems such as 

the build-up-of noise and signal distortion during processing. It may be modeled in the form of 

multidimensional systems. 

Image processing mainly include the following steps: 

 

 

 

Importing the image via image acquisition tools; 

Analysing and manipulating the image; 

Output in which result can be altered image or a report which is based on analysing 

that image. 

Some techniques which are used in digital image processing include: 

 

 

 

 

 

 

Image editing 

Image restoration 

Independent component analysis 

Linear filtering 

Partial different equations 

Pixelation 

97

APPLICATION- OVERVIEW 

The Brain Tumor is affecting many people worldwide. It is not only limited with the old 

age people but also detected in the early age. Brain tumor is the abnormal growth of cell inside the 

brain cranium which limits the functioning of the brain. Early detection the brain tumor is possible 

with the advancement of machine learning and image processing. In medical image processing is 

the most challenging and emerging field today. This describes the methodology of detection & 

extraction of brain tumor from pattern’s MRI scan image of the brain. This method incorporates 

with some noise removal functions, segmentation and morphological operations which are the basic 

concept of image processing. Detection and extraction of tumor from MRI scan images of the brain 

is done by using MATLAB software. 

MRI Imaging plays an important role in brain tumor for analysis, diagnosis and treatment 

planning. It’s helpful to doctor for determine the previous steps of brain tumor. Brain tumor 

detections are using MRI images is a challenging task, because the complex structure of the brain. 

Brain tumor can be detected by begin or malignant type. The benign being non-cancerous and 

malignant is cancerous. Malignant tumour is classified into two types; primary and secondary 

tumour benign tumor is less harmful than malignant. 

The basic idea is to develop application software to detect the presence of brain tumor in 

MRI images. We are using image processing techniques to detect exact position of tumor. 

TECHNOLOGY-USED 

There are various medical imaging techniques like x-ray, computed tomography, positron 

emission tomography, magnetic resonance imaging, are available for tumor detection. The MRI is 

the most commonly used modality for brain tumor growth imaging and location detection due to its 

higher resolution. 

1) To improve the performance and reduce the complexity involves in the image 

processing, we have investigated Berkeley wavelet transformation based brain tumor segmentation. 

2) To improve the accuracy and quality rate of the support vector machine based classifier, 

relevant features are extracted from each segmented tissue. 

98

Methodology 

This has two stages, first is pre-processing of given MRI image and after that segmentation and 

then morphological operations. 

• Input image 

• Multiparameter Calculations 

• Segmentation o brain tumor using Region of Interest command 

Image Processing Techniques 

Median Filtering for Noise Removal 

Is a non-linear filtering technique used for noise removal. It is used to remove salt and pepper 

noise from the converted gray scale image. It replaces the intensity values of the center pixel with 

the median of the intensity values in the neighbourhood of that pixel. Median filters are particularly 

effective in the presence of impulse noise. 

Various De-noising Filters 

Mean Filter- Based on average value of pixels 

Median Filter – Based on the median value of pixels 

Wiener Filter – Based on inverse filtering in frequency 

Hybrid Filter – Combination of median and wiener filter 

Modified hybrid median filter – Combination of mean and median Filter 

Morphology Based De-noising- Based on Morphological opening and closing Operations. 

Image Enhancement 

Poor contrast is one of the defects found in acquired image. The effect of that defect has great 

impact on the contrast of image. When contrast is poor the contrast enhancement method plays an 

important role. The gray level of each pixel is scaled to improve the contrast. Contrast 

enhancements improve the visualization of the MRI image. 

Edge Detection 

Is an image processing technique for finding the boundaries of objects with in images. It works 

by detecting discontinuities in brightness. It is used for image segmentation and data extraction in 

areas such as image processing, computer vision, and machine vision. 

99

Threshold 

It is a simple, effective, way of partitioning an image into a foreground and background. This 

image analysis technique is a type of image segmentation that isolates objects by converting gray 

scale images into binary images. Image threshold is most effective in images with high levels of 

contrast. 

Morphological Operation 

It is used as an image processing tools for sharpening the regions. It is a collection of non-linear 

operations related to the shape or morphology of features in an image. 

Segmentation 

Image segmentation is the process of portioning a digital image into multiple segments. It is 

typically used to locate objects and boundaries in image, the process of assigning a label to every 

pixel in an image such that pixels with the same level share certain visual characteristics. 

Brain Tumor Technology: 

• Classification Technology 

• Clustering Technology 

• Atlas-based segmentation 

• Histogram thersholding 

• Watershed and Edge detection in HSV colour model 

CONCLUSION 

The most of the existing methods had ignored the poor quality images like images with 

noise or poor brightness. Most of the existing work on tumor detection has neglected the use of 

image processing. To enhance the tumor detection rate further we will also integrated the new 

object based tumor detection. The proposed technique will have the ability to produce effective 

results even in case of high density of the noise. 

In this study Digital Image Processing Techniques are important for brain tumor 

detection by MRI images. The processing techniques include different methods like Filtering, 

Contrast enhancement, Edge detection is used for image smoothing. The pre-processed image are 

used for post processing operations like; threshold, histogram, segmentation and morphological 

which is used to enhance the images. 

100

REFERENCES 

1) Digital Image Processing (3 rd Edition): Rafael C. Gonzalez. Richard E. Woods 

2) Website: www.ijettcs.org 

3) Sharma, Komal, AkwinderKaur, and ShruthiGujral. “A review on various brain tumor 

detection techniques in brain MRI images.” IOSR Journal of Engineering Volume 04, Issue 05, pp: 

06-12,May 2014 

101

VIRTUAL REALITYANDITS TECHNOLOGIES 

M.Shree Soundarya 


ABSTRACT 

Virtual reality (VR) is a technology which allows a user to interact with a computersimulated 

environment, whether that environment is a simulation of the real world or an imaginary 

world. It is the key to experiencing, feeling and touching the past, present and the future. It is the 

medium of creating our own world, our own customized reality. It could range from creating a 

video game to having a virtual stroll around the universe, from walking through our own dream 

house to experiencing a walk on an alien planet. With virtual reality, we can experience the most 

intimidating and gruelling situations by playing safe and with a learning perspective. Very few 

people, however, really know what VR is, what its basic principles and its open problems are. In 

this paper a virtual reality and its basic technologies are listed and how VR works on it are 

described. 

INTRODUCTION 

Virtual realitymeans comes from both ‘virtual’ as a near us experience, and ‘reality’, 

something we can experience as human beings. The term itself can apply to almost anything that 

is possible to exist in reality but is stimulated by a computer. 

OBJECTIVES 

 

 

 

To know about Virtual Reality techand its works. 

To know the disparities between Virtual Reality(VR) and Augmented Reality(AR). 

To know about challenges in Virtual Reality with Virtual Entertainment(VE). 

SCOPE 

How it will be in future VR as a significant technology asworld-wide. 

102

VIRTUAL REALITY TECH AND WORKINGS 

A virtual experience includes three-dimensional images which appear life-sized to the 

user.Experiences are delivered to our senses via a computer and a screen, or screens, from our 

device or booth. A tool, or machine like a booth, or haptic system, may give our senses additional 

experiences, such as touch or movement. Speakers or earphones built into the headset, or set into a 

machine, will provide sound.For virtual reality to work, we have to believe in the 

experience. Virtual reality applications are getting closer and closer to a real-world background all 

the time. 

The technology needed for a virtual experience depends on the audience, purpose and of course the 

price point. If you are developing an interactive training simulator for a workplace, you will be 

looking for far more advanced technology. For everyday use here’s a quick overview of the 

technology and equipment are below. 

103

TABLE 1 

PARTICULARS 

EQUIPMENTS 

1 Virtual reality for smartphones. 

2 

3 

4 

Standalone virtual reality headsets. 

Powerful PC, laptop or console. 

Virtual reality headset. 

DISPARITIES BETWEEN VR AND AR 

Augmented reality and virtual reality are inverse reflections of one in another with what each 

technology seeks to accomplish and deliver for the user. Virtual reality offers a digital recreation of 

a real life setting, while augmented reality delivers virtual elements as an overlay to the real world. 

104

TABLE 2 

PARTICULAR DIFFERS VIRTUAL 

REALITY 

AUGMENTED 

REALITY 

1 PURPOSE It creates own It 

enhances 

reality that is experiences by adding 

completely computer 

generated and driven. 

virtual 

components such as 

digital images, graphics, 

or 

sensations as a new layer 

of interaction with the 

real world. 

105

2 DELIVERY 

METHOD 

It is usually delivered to 

the user through a headmounted, 

or hand- held 

controller. 

It is being used more 

and more in mobile 

devices such as laptops, 

smart phones, and 

tablets to change how 

the real world. 

CHALLENGES IN VR WITH VE 

Terminology in the virtual reality sector is fast changing. It’s argued that the first virtual 

reality devices emerged decades ago, but the experience they provided is entirely different to the 

virtual experience of today. As virtual reality becomes more intertwined with our daily lives, we 

start to think in terms of virtual environment. We can also define virtual enjoyment, which 

includes any virtual experience provided for pure entertainment – gaming, movies, videos and 

social skills for example – leaving more serious elements of virtual reality including training, 

education, and healthcare applications. Understanding how virtual reality works is crucial to being 

able to acknowledge this powerful technology honestly. It provides each of our senses with 

information to immerse us in a virtual experience, a near real experience – which our minds and 

bodies almost completely perceive as real. 

106

The Big challenges in the field of virtual reality are developing better tracking systems, finding 

more natural ways to allow users to interact within a virtual environment and decreasing the time it 

takes to build virtual spaces. While there are a few tracking system companies that have been 

around since the earliest days of virtual reality. Likewise, there aren’t many companies that are 

working on input devices specifically for VR applications. Most VR developers have to rely on and 

adapt technology originally meant for another discipline, and they have to hope that the company 

producing the technology stays in business. As for creating virtual worlds, it can take a long time to 

create a convincing virtual environment - the more realistic the environment, the longer it takes to 

make it. 

FUTURE ENHANCEMENT 

The future of Virtual Reality depends on the existence of systems that address issues of ‘large 

scale’ virtual environments. In the coming years, as more research is done we are bound to see VR 

become as mainstay in our homes and at work. As the computers become faster, they will be able to 

create more realistic graphic images to simulate reality better. It will be interesting to see how it 

enhances artificial reality in the years to come. It is very possible that in the future we will be 

communicating with virtual phones. Nippon Telephone and Telegraph (NTT) in Japan are 

developing a system which will allow one person to see a 3D image of the other using VR 

techniques. The future is virtual reality, and its benefits will remain immeasurable. 

107

CONCLUSION 

Virtual Reality is now involved everywhere. You can’t imagine your life without the use 

of VR Technology. Now we use mail or conference for communication while the person is not 

sitting with you, but due to technology distance is not matter.This technology give enormous scope 

to explore the world of 3D and your own imagination. 

108

STUDY ON REMOTE SENSING IN DIGITAL IMAGE PROCESSING 

V.Selvalakshmi &K.Sivapriya 


ABSTRACT 

Remote sensing is the acquisition of information about an object or phenomenon without making 

contact with the object and thus in contrast to on-site observation. Remote sensing is used in 

numerous fields, including geography, land surveying and most Earth Science disciplines; it also has 

military, intelligence, commercial, economic, planning, and humanitarian applications.in current 

usage, the term “remote sensing” generally refers to the use of satellite or aircraft based sensor 

technologies to detect and classify objects on Earth. 

Remote sensing image processing in nowadays a mature research area. The techniques 

developed in the field allow many real-life applications with great societal value. For instance, urban 

monitoring, fire detection or flood prediction can have a great impact on economical and 

environment issues. To attain such objectives, the remote sensing community has turned into a 

multidisciplinary field of science that embraces physics, signal theory, computer science, electronics 

and communications. Form a machine learning and signal/ image processing point of view, all the 

applications are tackled under specific formalisms, such as classification and clustering, regression 

and function approximation, image coding, restoration and enhancement, source un mixing, data 

fusion or feature selection and extraction. This paper serves as a survey of methods as applications, 

and reviews the last methodological advances in remote sensing and image processing. 

109

INTRODUCTION 

The Digital Image Processing is refers to processing digital images by means of digital 

computer. Note that a digital image is composed of a finite number of elements, each of which has 

a particular location and value. These elements are called picture elements, image elements, pels, 

and pixels. Digital image processing focuses on two major tasks improvement of pictorial 

information for human interpretation, processing of image data for storage, transmission and 

representation for autonomous machine perception. Some argument about where image processing 

ends and fields such as image analysis and computer vision start. The computer vision can be 

broken up into low-mid and high-level processes. Digital image processing helps us enhance 

images to make them visually pleasing, or accentuate regions or features of an image to better 

represent the content. For example, we may wish to enhance the brightness and contrast to make a 

better print of a photograph, similar to popular photo-processing software. In a magnetic resonance 

image (MRI) of the brain, we may want to accentuate a certain region of image intensities to see 

certain parts of the brain. Image analysis and computer vision, which go beyond image processing, 

helps us to make decisions based on the contents of the image. 

APPLICATIONS IN REMOTE SENSING 

Remote Sensing is the science and art of acquiring information (spectral, spatial, and temporal) 

about immaterial, objects, area, or phenomenon, without coming into physical contact with the 

objects, or area, or phenomenon under investigation. Without direct contact, some means of 

transferring information through space must be utilized. In practice, remote sensing is the stand-off 

collection through the use of a variety of devices for gathering information on a given object or 

area. In remote sensing, information transfer is accomplished by use of electromagnetic radiation 

(EMR). EMR is a form of energy that reveals its presence by the observable effects it produces 

when it strikes the matter. EMR is considered to span the spectrum of wavelengths from 10-10 mm 

to cosmic rays up to 1010 nm, the broadcast wavelengths, which extends from 0.30-15 mm.Sensors 

that collect information some remote distance from the subject. This process is called remote 

sensing of the environment. The remote sensor data can be stored in an analog format or in digital 

format the analog and digital remote sensor data can be analyzed using analog and/or digital image 

processing techniques. 

110

Scientists have made significant advances in digital image processing of remotely sensed data for 

scientific visualization and hypothesis testing and neural network image analysis, hyper spectral data 

analysis, and change detection. 

TECHNOLOGY USED IN REMOTE SENSING 

Since the beginning of the space age, a remarkable progress has been made in utilizing remote 

sensing data to describe study, monitor and model the earth’s surface and interior. Improvements in 

sensor technology, especially in the spatial, spectra, radiometric and temporal resolution, have 

enabled the scientific community to operationalise the methodology. The trend of development of 

remote sensing is being from panchromatic, multi-spectral, hyper-spectral to ultra-spectral with the 

increase in spectral resolution. On the other hand, spatial resolution is reaching its highest side of 

one metre resolution. 

A remote sensing application is a software application that processes remote sensing data. 

Remote sensing applications are similar to graphics software, but they enable generating geographic 

information from satellite and airborne sensor data. Remote sensing applications read specialized 

file formats that contain sensor image data, georeferencing information, and sensor metadata. Some 

of the more popular remote sensing file formats include: GeoTIFF, NITF, JPEG 2000, ECW,MrSID, 

HDF, and NetCDF. 

Remote sensing applications perform many features including: 

Change detection- determines the changes from images taken at different times of the same area. 

Othorectification-warp and image to its location on the earth. 

Spectral analysis-for example, using non-visible parts of the electromagnetic spectrum to 

determine if a forest is healthy 

Image classification- categorization of pixels based on reflectance into different land cover classes 

Until recently, the 2om spatial resolution of SPOT was regarded as ‘high spatial resolution’. Since the 

launch of IKONOS in 1999 a new generation of very high spatial resolution (VHR) satellites was 

born, followed byOuick Bird late 2001. The widely used Landsat and Spot sensore are now called 

‘medium resolution’. Especially the new satellite sensor generation meets the strong market demands 

from end-users, who are interested in image resolution that will help them observe and monitor their 

specific objects of interest. The increasing variety of satellites and sensors and spatial resolutions lead 

111

to a broader spectrum of application but not automatically to better results. The enormous amounts of 

data create a strong need for new methods to exploit these data efficiently. Airborne bases sensors to 

collect information about a given object or area. Remote sensing data collection methods can be 

passive or active. Passive sensors detect natural radiation that is emitted or reflected by the object or 

area being observed. In active remote sensing energy is emitted and the resultant signal that is 

reflected back is measured. 

Some latest powerful practical applications include anti-terrorism. 

 

 

 

Surgical strikes 

Disaster relief 

Detecting source of pollution 

112

CONCLUSION 

Digital Image Processing of satellite data can be primarily grouped into three categories: 

image rectification and restoration. Enhancement and information extraction. Image rectification 

is the pre-processing of satellite data for geometric and radiometric connections. Enhancement is 

applied to image data in order to effectively display data for subsequent visual interpretation. 

Information extraction is based on digital classification and is used for generating digital thematic 

map. 

REFERENCE 

1. Digital image processing (3 rd Edition): Rafael C. Gonzalez. Richard E. woods 

2. Introductory digital image processing (3 rd Edition): John R. Jensen 

3. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3841891/#! Po=16.0550 

4. https://www.geospatialworld.net 

113

BIG DATA ANALYTICS IN HEALTHCARE- A SURVEY 

Thirumalai & Thievekan 


ABSTRACT 

Like Oxygen, the world is encompassed by information today. The amount of 

information that we reap and eat up is flourishing forcefully in the digitized world. Expanding 

utilization of new advancements and online networking produce immense measure of information 

that can acquire marvelous data if legitimately broke down. This expansive dataset for the most part 

known as large information, don't fit in conventional databases in view of its rich size. Associations 

need to oversee and investigate enormous information for better basic leadership and results. Along 

these lines, huge information examination is accepting a lot of consideration today. In human 

services, huge information investigation has the likelihood of cutting edge quiet care and clinical 

choice help. In this paper, we audit the foundation and the different strategies for huge information 

investigation in human services. This paper likewise explains different stages and calculations for 

enormous information investigation and talk on its favorable circumstances and difficulties. This 

review ends up with a talk of difficulties and future bearings Keywords: big data, cloud computing, 

hadoop , big data mining, predictive analytics. 

INTRODUCTION 

The new advances in Information Technology(IT) manual for smooth making of 

information. For example, 72 long periods of recordings are transferred to YouTube 

consistently[26]. Social insurance part likewise has delivered tremendous measure of information 

by keeping up records and patient care. Opposite of putting away information in printed shape, the 

form is digitizing those boundless information. Those digitized information can be utilized to move 

forward the social insurance conveyance quality in the meantime diminishing the expenses and 

hold the guarantee of supporting a wide range of restorative and medicinal services capacities. 

Additionally it can give progressed customized mind, enhances understanding results furthermore, 

stays away from superfluous expenses. 

114

By depiction, huge information in human services alludes to electronic wellbeing datasets so 

extensive and complex that they are hard to deal with customary programming, equipment, 

information administration apparatuses and strategies Medicinal services huge information 

incorporates the clinical information, specialist's composed notes and remedies, therapeutic pictures 

for example, CT and MRI examines results, research center records, drugstore records, protection 

documents and other managerial information, electronic patient records (EPR) information; web 

based life posts, for example, tweets, reports on site pages what's more, various measure of 

therapeutic diaries. Thus, enormous measure of social insurance information are accessible for 

enormous information researchers. By understanding stencils and patterns inside the information, 

huge information examination appears to enhance mind, spare lives what's more, diminish costs. 

Consequently, enormous information investigation applications in human services exploit removing 

bits of knowledge from information for better choices making reason. Examination of huge 

information is the way toward reviewing tremendous measure of information, from various 

information sources and in different arrangements, to convey bits of knowledge that can empower 

basic leadership in genuine time. Different scientific ideas, for example, information mining and 

computerized reasoning can be connected to break down the information. 

Huge information systematic methodologies can be utilized to perceive inconsistencies which 

can be found because of incorporating immense measures of information from various 

informational indexes. information for better choices making reason. Investigation of huge 

information is the way toward reviewing colossal measure of information, from various information 

sources and in different organizations, to convey experiences that can empower basic leadership in 

genuine time. 

Different scientific ideas, for example, information mining and computerized reasoning can be 

connected to break down the information. Huge information investigative methodologies can be 

utilized to perceive peculiarities which can be found because of coordinating immense measures of 

information from various informational indexes. In whatever remains of this paper, initially we 

present the regular foundation, definitions and properties of huge information. At that point 

different huge information stages and calculations are talked about. In the long run, the difficulties, 

future bearings and conclusions are introduced. 

115

Definition and properties 

Amir Gandomi defined the 3V’s as takes after. Volume means the heaviness of the 

information. For the most part huge information sizes would be in terabytes or peta bytes or 

exabytes. Doug Beaver said that Facebook by and by stores 260 billion pictures which are around 

20 petabytes in size and it forms in excess of 1 million photographs every second. Assortment 

alludes the auxiliary decent variety in a dataset. Because of mechanical development, we can utilize 

diverse sorts of information those have different arrangement. Despite the fact that enormous 

information has been seen generally, it still has distinctive purpose of perspectives about its 

definition. Huge information in medicinal services is rising not just in view of its volume yet in 

addition the assortment of information writes and the speed at which it ought to be overseen. The 

accompanying definitions can assist us with understanding better the enormous information idea. 

Truth be told, enormous information has been characterized since 2001 itself. Certainly the 

measure is the significant ascribe that strikes a chord at whatever point we consider huge 

information. Nonetheless, it has some other properties likewise. Doug Laney (Gartner) 

characterized huge information with a 3V's model. 

It discussed the expansion of volume, speed and assortment. Apache Hadoop (2010) 

characterized enormous information as "datasets which couldn't be caught, overseen and handled 

by general PCs inside an worthy degree". In 2012, it was reclassified by Gartner as:"Enormous 

information is high volume, high speed and high assortment data resources that require new type of 

preparing to empower upgraded basic leadership, understanding disclosure and process 

streamlining Such information writes comprise of sound, video, content, picture, log documents et 

cetera. Enormous information design is arranged into three. They are organized, unstructured and 

semi-organized information. 

It is appeared in the accompanying figure. 

Figure-1. Content formats of big data. 

116

Structured data denotes the tabular data in spreadsheets and databases. The image, audio, 

video are unstructured data that are noted as difficult to analyze.Interestingly, nowadays 90% of big 

data are unstructured data. The size of this data goes on rising through the use of internet and smart 

phones. The characteristics of semistructured data lie between structured and unstructured data. It 

does not follow any strict standards. 

XML (Extensible Markup Language) is a common example of semi-structured data. The third 

‘V’ velocity means the speed at which data is produced and analyzed. As mentioned earlier, the 

emergence of digital devices such as smart phones and sensors has allowed us to create these 

formats of data in an extraordinary rate. The various platforms behind big data and algorithms are 

discussed in detail in the later sections 

RELATED TECHNOLOGIES 

A. Big data platforms 

As in ,huge information utilizes conveyed capacity innovation in light of distributed 

computing as opposed to nearby capacity. Some enormous information cloud stages are Google 

cloud administrations, Amazon S3 and Microsoft Azure. Google's disseminated document 

framework GFS (Google File System) and its programming model Map lessen are the lead in the 

field. The execution of guide diminish has gotten a substantial measure of consideration in huge 

scale information handling. So numerous associations utilize enormous information handling 

structure with delineate. Hadoop, a powerful angle in huge information was produced by Yahoo 

and it is an open-source adaptation of GFS [29]. Hadoop empowers putting away and preparing 

huge information in dispersed condition on vast bunches of equipment .Colossal information 

stockpiling and quicker preparing are bolstered by hadoop. Hadoop Distributed File System 

(HDFS) gives dependable and versatile information stockpiling. HDFS influences different 

duplicates of every datum to square and disseminates them on frameworks on a bunch to empower 

dependable access. HDFS bolsters distributed computing through the utilization hadoop, a 

disseminated information preparing stage. 

117

Another, 'Huge Table' was produced by Google in 2006 that is utilized to process 

immense measure of organized information. It likewise bolsters outline. ystem. It is a versatile 

disseminated information store manufacturedfor Amazon's stage. It gives high unwavering quality 

cost viability, accessibility and execution. Tom white explains different instruments for huge 

information investigation. Hive, a structure for information warehousing over hadoop. It was 

worked at Facebook. Hive with hadoop for capacity and handling meets the adaptability needs and 

is costeffective. Hive utilizes an inquiry dialect called HiveQL which is similar on SQL. A scripting 

dialect for investigating extensive dataset is called 'Pig'1 

A conclusion of guide diminish is that written work of mappers and reducers, 

incorporating the bundle and code are extreme thus the advancement cycle is long. Subsequently 

working with mapreduce needs encounter. Pig survives this feedback by its effortlessness. It 

enables the designers to compose straightforward Pig Latin questions to process the huge 

information and consequently spare the time. A dispersed section situated database Hbase based 

over Hadoop Distributed Record System. It can be utilized when we require irregular access of 

expansive datasets. It accelerates the execution of activities. Hbase can be gotten to through 

application programming interfaces (APIs) like REST (Illustrative State Transfer) and java. Hbase 

does not have its own inquiries, so it relies upon Zookeeper. Zookeeper oversees tremendous 

measure of information. This permits dispersed procedure to oversee through a namespace of 

information registers.This dispersed administration additionally has ace and slave hubs like in 

hadoop. Another imperative instrument is Mahout. It is an information mining and machine 

learning library. It can be sorted as aggregate separating, classification, bunching and mining. It can 

be executed by Mapreduce in a dispersed mode. Enormous information investigation isn't just in 

view of stages yet additionally examination calculations plays a huge part. 

118

B. Algorithmic techniques 

Huge information mining is the strategy for winnowing shrouded, obscure however 

valuable data from enormous measure of information. This data can be utilized to anticipate f uture 

circumstances as an assistance to basic leadership process.Accommodating learning can be found 

by the utilization of information mining systems in social insurance applications like choice 

emotionally supportive network. The enormous information created by human services associations 

are extremely muddled and immense to be dealt with further more, investigated by regular 

techniques. Information mining stipends the methodology to change those groups of information 

into helpful data for choice help. Huge information mining in medicinal services is tied in with 

learning models to anticipate patients' ailment. For instance, information mining can encourage 

social insurance protection associations to distinguish scoundrels and abuse, medicinal services 

foundations settle on choices of client relationship administration, specialists distinguish successful 

medications and best practices, and patients get made strides furthermore, more conservative 

human services administrations. This prescient examination is generally utilized as a part of 

medicinal services. There are different information mining calculations talked about in 'Top 10 

calculations in information mining' by Wu X et al. It talked about assortment of calculations 

alongside their constraints. Those calculations include grouping, order, relapse, measurable 

realizing which are the issues in information mining examination. The ten calculations talked about 

incorporate C4.5, k-implies, Apriori, Support VectorMachines, Naïve Bayes, EM, CART, and so 

on. Enormous information investigation incorporates different techniques, for example, content 

examination, mixed media investigation et cetera. 

In any case, as given over, one of the urgent classifications is prescient investigation which 

incorporates different measurable strategies from demonstrating, information mining and machine 

discovering that break down present and verifiable certainties to make expectation about future. In 

healing center setting, there are prescient strategies used to distinguish on the off chance that 

somebody might be in danger for readmission or is on a genuine subsidence.This information 

encourages advisors to settle on essential care choices. Here it is important to think about machine 

learning since it is broadly utilized in prescient examination. The procedure of machine learning is 

especially indistinguishable of information mining. Them two chase through information to 

119

Figure-2. Machine learning algorithms - A hierarchal view. 

120

Corridor laid out a strategy for building taking in the tenets of the huge dataset. The 

approach is to have a solitary choice plan produced using a gigantic subset of information. 

In the interim Patil et al. sought after a cross breed way blending the two hereditary 

calculation and choice tree to make an propelled choice tree to enhance execution and 

effectiveness of calculation. With the expanding information in the territory of enormous 

information, the assortment of systems for dissecting information is spoken to in 

'Information Lessening Techniques for Large Qualitative Data Sets'. It depicts that the 

determination for the specific method is in view of the kind of dataset and the way the 

example are to be examined. Its connected K-implies bunching utilizing Apache Hadoop. 

They went for proficiently investigating substantial dataset in an insignificant measure of 

time. They likewise clarified that the exactness and discovery rate are influenced by the 

quantity of fields in log records or not Along these lines, their tried outcomes demonstrate 

the right number of groups and the right measure of passages in log documents, yet the rate 

of precision lessens when the quantity of sections increments. The outcome demonstrates 

that the precision should be improved .Classification is one of the information mining 

techniques used to anticipate and order the foreordained information for the particular class. 

There are diverse orders techniques propose by specialists. The generally utilized strategies 

are portrayed by Han et al. 

It incorporates the accompanying: 

Neural network algorithm 

Decision tree induction 

Bayesian classification 

Rule based classification 

Support vector machine 

K-Nearest neighbor classifier 

Rough set approach 

Genetic algorithm, 

Fuzzy set approach 

Any of the previously mentioned arrangement strategies can be connected to characterize the 

application arranged information. The material arrangement strategy is to be 

121

picked by the kind of use and the dimensionality of the information. It is a major test to the 

analysts to choose and apply the fitting information mining arrangement calculation for 

diagnosing medicinal related issues. Picking the right strategy is a testing errand. 

The correct technique can be picked simply in the wake of breaking down all the accessible 

arrangement techniques and checking its execution in term of precision. Different inquires about 

have been completed in the region of medicinal analyses by utilizing arrangement system. The 

most imperative reality in restorative finding framework is the exactness of the classifier. This 

exploration paper investigations the diverse characterization strategies connected in restorative 

conclusions and thinks about the execution of order precision. C4.5 is connected to break down 

the SEER dataset for bosom disease and order the patients either in thebeginning stage or pre 

malignancy organize The records broke down are 500 and the precision accomplished in testing 

datasets is 93% Shweta utilized the Naïve Bayes, ANN, C4.5 and choice tree calculations for 

analyze and visualizations of bosom tumor.The outcomes demonstrate that the choice trees give 

higher exactness of 93.62 % where Credulous Bayes gives 84.5%, ANN produces 86.5% and 

C4.5 produces 86.7% of exactness. Chaitrali is utilized Choice Trees, Naïve Bayes and Neural 

Network calculations for dissecting coronary illness. The outcomes examination tells that the 

Naïve Bayes accomplishes 90.74% of exactness while Neural Network and Decision Trees give 

100% and 99.62% of exactness separately. Diverse information mining systems were connected 

to foresee coronary illness The exactness of each calculation is checked and expressed as Naïve 

Bayes, Decision Tree and ANN are accomplished 86.53%, 89% and 85.53% of exactness 

individually. The three distinct information mining calculations, ANN, C4.5 and Decision Trees 

are utilized to investigate heart related infections by utilizing ECG signals . The examination 

comes about plainly demonstrate the Decision Tree calculation performs best and gives the 

exactness of 97.5%. C4.5 calculation gives 99.20% exactness while Naïve Bayes calculation 

gives 89.60 % of exactness.Here these calculations are utilized to appraise the supervision of 

liver scatter. Christobel et al. connected KNN technique to diabetic dataset. It gives the exactness 

of 71.94% with 10 overlap cross approval. C5.0 is the arrangement calculation which is 

appropriate for enormous informational collections. It defeats C4.5 on the speed, memory and the 

execution. C5.0 technique works by part the example in light of the field that gives the most 

extreme data pick up. 

The C5.0 framework can part tests with respect to of the greatest data pick up field. The example 

subset that is got from the past part will be part later. The activity will proceed until the example 

122

subset can't be part and is for the most part as per another field. At long last, think about the least 

level split, those example subsets that don't have prominent commitment to the model will be 

dropped. C5.0 approach effortlessly handles the multi esteem trait and missing quality from 

informational collection .The C5.0 manage sets have recognizably brings down mistake rates on 

inconspicuous cases for the rest and woods datasets. The C4.5 what's more, C5.0 govern sets have 

the same prescient precision for the pay dataset, yet the C5.0 run set is littler. The times are nearly 

not similar. For example, C4.5 required about 15 hours finding the run set for woods, yet C5.0 

finished the errand in 2.5 minutes. C5.0 generally utilizes a request of extent less memory than 

C4.5 amid administer set development [20]. So obviously C5.0 approach is superior to C4.5 in 

numerous viewpoints. Hsi-Jen Chiang proposed a strategy for breaking down prognostic pointers 

in dental embed treatment. They break down 1161 inserts from 513 patients. 

Information on 23 things are taken as effect factors on dental inserts. These 1161 inserts are 

examined utilizing C5.0 technique. Here 25 hubs are created by utilizing C5.0 approach. This 

model accomplishes the execution of 97.67% precision and 99.15% of specificity. 

CHALLENGES AND FUTURE DIRECTION 

Enormous information investigation not just gives enchanting openings yet in addition 

faces parcel of difficulties. The challenge begins from picking the huge information examination 

stage. While picking the stage, a few criteria liken accessibility, usability, adaptability, level of 

security and congruity ought to be thought about Alternate difficulties of enormous information 

investigation are information inadequacy, adaptability furthermore, security. Since distributed 

computing plays a real part in huge information investigation, cloud security ought to be 

considered. Studies demonstrate that 90% of huge information are unstructured information. 

Yet, the portrayal, investigation and access of various unstructured information are as yet a test. 

Information convenience is additionally basic in different human services regions like clinical 

choice help for settling on choices or giving data that advisers for take choices. Huge information 

can settle on choice help more straight forward, quicker and that's only the tip of the iceberg 

exact on the grounds that choices depend on higher volumes of information that are more 

present and significant. This needs versatile investigation calculations to create opportune 

outcomes Be that as it may, the vast majority of the present calculations are wasteful in terms of 

huge information examination. So the accessibility of viable examination calculations is 

additionally important. Worries about protection and security are prevalent, despite the fact that 

123

these are progressively being endeavored by new validation methodologies and strategies that 

better secure patient identifiable information. 

CONCLUSION 

A lot of heterogeneous medicinal information have turned out to be accessible in different 

human services associations. The rate of electronic wellbeing record (EHR) appropriation keeps 

on moving in both inpatient and outpatient viewpoints. Those information could be an 

empowering asset for inferring bits of knowledge for enhancing tolerant care and lessening waste 

Investigating the gigantic measure of medicinal services data that is recently accessible in 

computerized configuration should empower propelled recognition of great treatment, better 

clinical choice help and precise forecasts of who is likely to become ill. This requires elite 

processing stages and calculations. This paper surveys the different enormous information 

investigation stages and calculations and difficulties are examined. In view of the examination, 

albeit restorative analyze applications utilize distinctive calculations, C4.5 calculation gives better 

execution. Yet at the same time the act of spontaneity of C4.5 calculation is required to boost 

exactness, handle huge measure of information, decrease the space prerequisite for huge measure 

of datasets and bolster new information composes and to diminish the mistake rate. C5.0 

approach conquers these reactions by creating more precision; requiring less space when volume 

of information is expanded from thousands to millions or billions. It additionally has bring down 

mistake rate and limits the prescient blunder. C5.0 calculation is the conceivably appropriate 

calculation for any sort of therapeutic judgments. 

In the event of enormous information, the C5.0 calculation works quicker and gives the 

better exactness with less memory utilization. Despite the restricted work done on enormous 

information examination up until now, much exertion is expected to beat its issues identified with 

the previously mentioned challenges. Likewise the fast advances in stages and calculations can 

quicken the execution. 

124

REFERENCES 

1. Alexandros Labrinidis and H. V. Jagadish, “Challenges and opportunities with big data,” Proc. 

VLDB Endow. 5, pp. 2032-2033, August 2012. 

2. Aneeshkumar, A.S. and C.J. Venkateswaran, “Estimating the surveillance of liver disorder 

using classification algorithms”. Int. J. Comput. Applic., 57: pp. 39-42, 2012. 

3. Amir Gandomi, Murtaza Haider, “Beyond the hype: Big data concepts, methods, and 

analytics,” International Journal of Information Management 35, pp. 137-144, 2015 

4. Chaitrali, S., D. Sulabha and S. Apte, “Improved study of heart disease prediction system using 

data mining classification techniques," Int. J. Comput. Applic. 47: 44-48, 2012. 

5. Doug Beaver, Sanjeev Kumar, Harry C. Li, Jason Sobel, Peter Vajgel, Facebook Inc, “Finding 

a Needle in Haystack: Facebook’s Photo Storage” 2010. 

6. Fay Chang, Jeffrey Dean, Sanjay Ghemawat, Wilson C. Hsieh, Deborah A. Wallach, Mike 

Burrows, Tushar Chandra, Andrew Fikes, and Robert E. Gruber, “Bigtable: A Distributed Storage 

System for Structured Data,” ACM Trans. Comput. Syst. 26, 2, Article 4, June 2008 

7. Jason Brownlee, “Machine Learning Foundations, Master the definitions and concepts”, 

Machine Learning Mastery, 2011. 

8. Rulequest research, http://rulequest.com/see5- comparison.html. 

9. Ghemawat, Howard Gobioff, and Shun-Tak Leung: “The Google file system,” Proceedings of 

the 19th ACM Symposium on Operating Systems Principles (SOSP), Bolton Landing, NY, USA, 

October 19-22, 2003. 

125

combinepdf (5)

Create successful ePaper yourself

Delete template?

Save as template?