An evolutionary algorithm to perform set matching is presented. A set matching between two sets is a transformation, which when applied to one of the sets brings it into the closest match with the other set. Two measures of similarity between sets, Hamming distance and Hausdorff distance, are used to determine how well a transformation matches the sets. We focus on the case in which the sets are subsets of I 2 = [0, 1] × [0, 1]; such sets can be interpreted as black and white images. The space of transformations to find the matching is the space of affine transformations on I 2.
