An Intelligent Novel Model for an Embedded Stereo Vision System

This paper proposes a novel algorithm and hardware architecture for an embedded stereo vision system which is appropriate for robotics applications. In this paper we distributed this process to three re-programmable processor unit as pre-processing, post-processing and stereo matching. In order to accomplish with on two Digital Signal Processors (DSPs) for the pre and post processing as well as a Field Programmable Gate Array (FPGA) for stereo matching and describe our reasons to use this architecture for robotic applications. Hence we described how to implement this algorithm to achieve the utmost speed for stereo matching operation.