Image Enhancement - Nubacad.com

Image Enhancement
Modify color or gray scale of individual pixels in an image.
The main objective of image enhancement is to process an
image so that the result is more suitable than the original
image for specific application.
Image Enhancement Methods fall into two broad categories:
Spatial domain (image plane)
Pixel processing (direct manipulation of pixels)
Gray level transformation: Data independent
Histogram processing: Data-dependent
Arithmetic operations
Spatial filtering
Frequency domain
Fourier transform
Filtering
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 1

Spatial Domain
Spatial domain processes denoted as:
g (x,y) = T [f (x,y)];
where f (x,y) is the input image, g (x,y) is the processed image, T is
the operator on f, defined over some neighborhood of (x,y).
Gray Level Trans. Func.
The simplest form of T is when
the neighborhood size is 1x1
(that is a single pixel). (Point
Processing)
s = T(r)
r and s denote the gray level of f
(x,y) and g (x,y) at point (x,y),
respectively.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 2

Gray Level Transformation Function: s = T(r)
The effect of gray level transformation of Fig. 3.2 would be to
produce an image of higher contrast than the original by
darkening the levels below m and brightening the levels above
m in the original image. This technique is known as contrast
stretching.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 3

Basic Gray-level Transforms
• Operated on individual pixel’s intensity values:
s = T(r). r: original intensity, s: new intensity
• Data independent pixel-based enhancement
method.
• Approaches
– Image negatives
– Log transform
– Power law transform
– Piece-wise linear transform
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 4

Image Negatives
• s = T(r) = L-1-r
• Similar to photo negatives.
• Suitable for enhancing gray details in dark background.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 5

Log Gray-level Transform
• s = T(r) = c log(1+r) • expand dark value to enhance
details of dark area
(a) Fourier spectrum and (b) Result of applying
Log transform.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 6

Piece-wise Linear Gray-level Transform
• Allow more control on the complexity of T(r).
– Contrast stretching
– Gray-level slicing
– Bit-plane slicing
Contrast stretching
Low contrast images can result poor
illumination, lack of dynamic range in the
imaging sensor, or even wrong setting of
imaging aperture during image acquisition.
The idea behind contrast stretching is to
increase the dynamic range of the gray
levels in the image behind process.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 7

Contrast Stretching
Run program spatial_c.m
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 8

Gray-level Slicing
Provides a binary image
Gray level slicing is
accomplished by
highlighting a specific
range of gray levels in
an image by displaying
a high value for all gray
levels in the range of
interest and a low value
for all other gray levels.
Applications:
Enhancing features
such as masses of water
in satellite imagery and
enhancing flaws in x-
ray images.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 9

Bit-Slicing
Provides highlighting
the contribution made
to total image
appearance by specific
bits.
Suppose that each
pixel in an image is
represented by 8 bits.
Imagine that the
image is composed of
eight 1-bit planes,
ranging from bit plane
0 for the least
significant bit to bit
plane 7 for the most
significant bit.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 10

Histogram
• The histogram of an image records the
frequency distribution of gray levels in
that image.
• The histogram of a digital image with
gray levels in the rage [0, L-1] is the
discrete function h(r k ) = n k where r k is
the k-th gray level and n k is the number
of pixels in the image having gray level
r k .
• Normalized histogram
p r (r k ) = n k /n, for k = 0,1,…, L-1
• p(r k ) gives the estimates of the
probability of occurrence of gray level
r k .
• A plot p(r k ) versus r k is called the
Histogram
• Several effects of histograms are shown
at the right side.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 11

Cumulative Histogram
Records the cumulative
frequency distribution of
gray levels in an image.
k
nk
sk
= T ( r)
= ∑
n
j=
0
The cumulative frequency of a gray
level, i, is the number of times that a
gray level less than or equal to i
occurs in an image.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 12

Histogram Equalization
Redistribution of gray levels: an attempt to
flatten the frequency distribution.
The probability of occurrence of gray level r k ,
p r (r k ) = n k /n, for k = 0,1,…, L-1;
where n is the total number of pixels, r k is the
k-th gray level and n k is the number of pixels
having gray level r k .
Then the Histogram Equalization or Histogram
Linearization transformation function:
k
k
n
j
sk
= T ( rk
) = ∑ pr
( rj
) = ∑ ;
n
j=
0
j=
0
for k = 0,1,2,..,L-1
The function s can be considered as cumulative
sum of p r (r k )
More gray levels are allocated
where there are most pixels,
fewer gray levels where there
fewer pixels.
This tends to increase contrast in
the most heavily populated
regions of the histogram.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 13

Histogram Equalization
Histogram equalization
becomes histogram
specification, if instead
of requiring a flat
histogram, we specify
a particular shape
explicitly.
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 14

Histogram Equalization …
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 15

Histogram Equalization …
• A gray-level transformation method that forces the transformed gray level to
spread over the entire intensity range.
– Data dependent,
– (usually) Contrast enhanced
• Usually, the discrete-valued histogram equalization algorithm does not yield
exact uniform distribution of histogram.
• In practice, one may prefer “histogram specification”
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 16

Enhancing using Arithmetic/Logic Operations
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 17

Enhancing using Arithmetic/Logic Operations
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 18

Image Subtraction
Subtraction of two images result in a new image whose pixel at coordinates
(x,y) is the difference between the pixels in that same
location in the two images being subtracted.
The difference between two images f(x,y) and h(x,y) is expresses by:
g(x,y)= f(x,y) - h(x,y)
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 19

Image Subtraction
May 07, 2011 Dr. Al-Amin http://www.geocities.com/alamin_bhuiyan 20