A Study Of Local Stereo Matching Algorithm In Binocular Stereo Vision

Stereo vision has always been a hot research topic in the field of computer vision.Stereo vision using two or more cameras shooting the same scene,it restores the original scene of the three-dimensional information from the two-dimensional image information according to the geometric principle.A stereo vision system consists of six parts,image acquisition,camera calibration,image preprocessing,stereo matching,3D reconstruction and post-processing.Stereo matching is important and has far-reaching significance in stereo vision.Binocular stereo matching is based on the principle of human vision,it is observed from two different perspectives of the same visual target.Binocular stereo matching matches the perceived image.Binocular stereo matching matches the perceived image.Binocular stereo matching is mainly used in the field of robot navigation,parameter detection of micro operating system,three-dimensional measurement and virtual reality.By comparing the accuracy and the processing speed,the article chooses the algorithm based on Census Transform as the stereo matching algorithm.This algorithm has good robustness,low computational complexity and it is easy to implement in the hardware.The article uses Matlab as the simulation platform.In order to improve the robustness and accuracy of stereo matching algorithm,this article proposed a stereo matching algorithm based on improved Census transform.Firstly,the gray value of central pixel is replaced by the average value of sub-window of minimum variance,which overcomes the dependence of neighborhood pixels on the center pixel;secondly,the adaptive support weights is used in the Census transform,and the accuracy of stereo matching is improved by increasing the information of pixels.In costs aggregation,the adaptive support weights is used again,according to epipolar constraint, the disparity of minimum cost is obtained by the method of WTA,which is the disparity of matching point.The disparity of initial disparity map is enhanced by means of left-right cheking and Sub-pixel Enhancement.At the last,the proposed algorithm is tested by using the stereo matching images of theMiddlebury website,then it is compared with the the current other good algorithms by experimental results and is analyzed in robustness and the accuracy.