Contrast analysis

This article describes contrast analysis as a method for image recognition and automatic classification in computer science . The contrast level is increased in the image in equal steps. Similar to the cluster analysis , it is possible to classify objects on the image.

algorithm

Contrast increase

Various algorithms from computer graphics can be used to increase the contrast . The aim is to reduce the color palette of the colors used in the picture. At the same time, the saturation of the image can be reduced in order not to unnecessarily increase the number of (known) target classes. The image can also be converted to grayscale at the beginning of the entire run. Each RGB color value is converted into the corresponding gray level (i.e. where ). While the change in saturation depends on the type of image, the increase in contrast must take place in a finite number of steps, evenly monotonously. ${\ displaystyle ({(r, g, b) | r, g, b \ in {0, ..., 255}})}$ ${\ displaystyle ({\ overline {g}}, {\ overline {g}}, {\ overline {g}})}$ ${\ displaystyle {\ overline {g}} = {\ frac {r + g + b} {3}}}$

A new run is started again and again until the picture can be assigned to a known class. The last stage of increasing the contrast is reached when the color palette only contains the basic colors of the color spectrum used. Then the algorithm has to terminate.

classification

the resulting pattern is compared with the known patterns
if there is no match, a new run of the contrast enhancement is started
as soon as an identical pattern is found, the target class is found
If no identical pattern is found up to the maximum contrast, the generated patterns are checked backwards for known similar classes, as soon as enough pixels are identical, the target class has been found

Contrast analysis

contents

algorithm

Contrast increase

classification

See also