Histogram Equalization

 

With this applet you can manually scale the image color in order to adjust the contrast and brightness.

How to use the applet

Six images are located in the applet. The first one is the input image followed by its histogram. The third image is the cumulative color distribution function. Below you see three images you can work with: the color scaled result, its histogram and its cumulative color distribution function.

There is a blue line drawn into the histogram of the input image. It represents the scaling function of the input image, which is equal to the identity function when the applet  starts. You can deform this function when you click on it with the mouse pointer and drag it over the image. Observe the images down left and right. The contrast and the cumulative function will be changed. Afterwards there are little black rectangles on the scale function which define interpolation points. You can let them disappear when you move the mouse pointer over such a black rectangle and press the left mouse button. In this way you can reconstruct the original image.

Of course you can change the input image by pressing the -button. A scaling function that manipulates the image in a way, that the histogram of the result image has a constant number of image points in constant intensity ranges, is called ‘histogram equalization’. You can proof that the cumulative distribution function performs that. When you press the -button the scale function will be converted into the cumulative distribution function automatically. To reconstruct the original image press the -button.

Downloads

SourceCode

JavaDoc


 

Authors: René Iser and Simon Winkelbach

Other Applets