Next: Remarks Up: The CMSC 426 Image Previous: Edge Detection

Template Matching

In template matching, you are given a template image which you want to try and locate/find in a one or more target images. This is done by match the template image to all (or many) of the possible places it could be located at in the target image. A distance function (typically a simple Euclidean distance) is applied to the match to measure the similarity of the template and the image at the location. The algorithm then picks the location with smallest distance as the location of the template image in the target image.

There are several ways in which you can vary this basic algorithm. The first is by using ``fancy'' distance functions. This may be necessary for images which have different overall brightnesses etc... Another set of modifications attempt to reduce the number of possible locations which are actually looked at. One such method is to use image pyramids. Another method is to only match every few pixels, and then for good matches, try matching for all the pixels in the neighborhood.

The implementation of template matching in the CMSC 426 Image Workshop is still incomplete.