A Novel Vision Modelling for Mobile Robots
The image segmentation of a robot binocular stereo vision system is the key issue in imaging processing. In this paper, the method of 2-D maximum entropy threshold image segmentation with chaos optimization algorithm is used to segment the images information collected by a robot vision system, and the algorithm is checked by a real robot binocular stereo vision system. Moreover, for a real environment simulation a new programming interface library for IEEE 1394 video devices on Linux is proposed. The simulation experiments of the proposed method in comparison to the best previous research are shown better efficiency in using space information of an image and shortening the calculation time.