Adaptive integrated image segmentation and object recognition

This paper presents a general approach to image segmentation and object rec ognition that can adapt the image segmentation algorithm parameters to the changing environmental conditions. Segmentation parameters are represented by a team of generalized stochastic learning automata and learned using con nectionist reinforcement learning techniques. The edge-border coincidence m easure is first used as reinforcement for segmentation evaluation to reduce computational expenses associated with model matching during the early sta ge of adaptation, This measure alone, however, can not reliably predict the outcome of object recognition, Therefore, it is used in conjunction with m odel matching where the matching confidence is used as a reinforcement sign al to provide optimal segmentation evaluation in a closed-loop abject recog nition system. The adaptation alternates between global and local segmentat ion processes in order to achieve optimal recognition performance. Results are presented for both indoor and outdoor color images where the performanc e improvement over time Is shown For both image segmentation and object rec ognition.