Real-time feature extraction from video stream data for stream ...

ai.cs.uni.dortmund.de

Real-time feature extraction from video stream data for stream ...

D. Implemented processors in the streams framework

Processors in Package stream.image

This package provides several processors for transforming images (video frames).

Processor BorderDetection

This processor performs a border detection on the input image. A pixel is said to be a

border pixel, if its’ color differs from the color of at least one of its’ four neighboring

pixels. By increasing the tolerance, the amount of neighboring pixels that have to have

the same color value can be decreased.

Parameter Type Description Required

output String The name/key of the output image is stored. If this name

equals the name of the input image, the input image is going

to be overwritten.

tolerance int The number of neighboring pixels that may have a different

color value, without causing, that the actual pixel becomes

recognized as a border pixel. The higher the tolerance is,

the less border pixels will be found.

image String The name of the attribute that contains the byte array data

of the image.

Table .4.: Parameters of class BorderDetection.

false

false

true

Processor ColorDiscretization

The processor discretizes the color space of the input image by discretizing each single

RGB color channel.

Parameter Type Description Required

bins Integer Set the number of discrete color values, each channel in

divided into.

output String The name/key under which the output image is stored. If

this name equals the name of the input image, the input

image is going to be overwritten.

image String The name of the attribute that contains the byte array data

of the image.

Table .5.: Parameters of class ColorDiscretization.

false

false

true

110

More magazines by this user
Similar magazines