Hough Transformation

 

This applet demonstrates the Hough transformation for lines. It  detects lines in the image space by finding clusters in the Hough space.

Theoretical background

How to use the applet

Below you can see two pictures. The first one is the input picture in which the lines will be searched. The second picture shows the Hough space.

When you move the mouse pointer over the input image a cyan colored line appeares on the Hough image. It represents the transformation line of the

active mouse position  which would be add into the Hough space if you draw a dot by pressing the left mouse button now. You can also drag the mouse

button to tranform  points. While moving the mouse over the Hough space a red line appears on the original image.It is the same effect as on the left

side but now it concerns the back transformation.

 There are two possible operators to work with. The first one is the simple point. Just click into the left image and see the transformation line in the Hough

space. In order to draw a perfect line, a line operator can be choosen. Both tools can be chosen by pressing the   -button.

If you want to find some maxima in the Hough space automatically just press the -button. Then a menu appears where you have to

choose the number of clusters which should be found. Afterwards the clusters are visualised as yellow lines in the image space and the positions of the

maxima are marked by green quadrats in Hough space. In this way you can internalise how the Hough transformation works.

For a better illustration of the pointcoordinates you can active coodinate systems  in both pictures. Just press the -button.

The last feature is the -button which cleans both images.

Downloads

SourceCode

JavaDoc