Distance Transform, maps binary images to distance from background
Fourier Transform, maps image into spatial frequency domain
Hough
Transform, maps image into votes for specified shapes.
Most of the image processing operators demonstrated in HIPR transform an input image to form a new image. However, the operators demonstrated in this section produce output images whose character is generally quite different from the character of the input images. This difference might be in the geometry of the information in the image or the nature of the information itself. Of course, the size of the output image might be quite different from that of the input image (and the values at each pixel might be different, as in the complex number output of the Fourier transform).
The usual purpose of applying a transformation is to help make more obvious or explicit some desired information. In the case of the operators in this group, the main effect is to make explicit:
The transformation is often followed by a thresholding operation, which is intended to select the most prominent or relevant features. It may then be possible to apply an inverse transform, for the purpose of reconstructing the geometry of the original image, except with the desired features explicit or enhanced.
©2003 R. Fisher, S. Perkins,
A. Walker and E. Wolfart.
