Dynamic Obstacle Detection Through Cooperation of Purposive Visual Modules of Color, Stereo, and Motion

In this paper we present a new framework for detection of dynamic obstacles in tire unstructured outdoor road environment by purposively irttegmting binocular color image sequence. In our system, color image segmentation, stereo obstacle detection, visual egomotion estirnation,and moving object analysis, are all built-in task-oriented modules and hence are eficient and robust. Most of these functions can be perfomied in realtime. They are activated and integrated adaptively in the manner of neural parallel distributed processing(PDP). Experimental results are given to validate our philosophy of so-called purposive vision , retinal mapping , adaptive integration and parallel distributed processing.