Hough Transformation in Normal Form


This kind of Hough transformation also detects lines in image space except that lines are represented in normal form. For further information concerning this representation see Mathmatical background

How to use the applet

In the applet shows two images. The first one on the left side is the input image. The right image shows the Hough space. Because of the normal form the coordinates of Hough space are the distance r to the origin along vertical axis and the angel Θ along horizontal axis. When you move the mouse pointer over the input image you can see the Hough representation of this point on the right image. It works in the same way when the mouse is moved over the Hough image. In this case you can observe the back transformation of the active position. If you have chosen the "empty" picture you can draw points in the input image by dragging the mouse (delete all points with ). You can change the input image by clicking the -button .The Hough image on the right side will be updated automatically. The applet can also find clusters automatically in Hough space. After pressing the -button a menu appears where you can choose the number of clusters, which are supposed to be found in the original image. The back transformed points appears as yellow lines on the left image. The found clusters are marked by green squares on the right hand side.





Authors: René Iser, Behrang Karimibabak and Simon Winkelbach

Other Applets