A genetic algorithm based method to improve image segmentation

Segmentation of textured images is becoming more and more important in applications, as quality control or remote sensing. Segmentation of textured images is demanding. A new genetic algorithm based method to post-process segmented texture images is presented. A genetic algorithm is used to extract web-like rules from segmented texture images. These rules are checked and they are used in post-processing to improve the segmentation. An unsupervised image segmentation and definition of classes by class prototypes are assumed. Some preliminary results are presented