Dense Disparity Algorithm Research Of Binocular Stereo Vision

Stereo vision is a research hotspot of the computer vision domain,it's objective is the restoration of three-dimensional structure of a scene based on geometry principle requiring two or more cameras to observe the identical scenery shown at a few different orientations. Along with the development of computer technology, stereo vision will be widely applied in the production automation,automatic navigation,the medicine,information security as well as antiterrorism domains, etc. Therefore the research of stereo has significance and the practical value. The thesis discusses the stereo matching of the computer vision.Trough analyzing the existed occlusion detecting algorithm and using the concept of the left-right line of sight, we introduce the new method of occlusion detection and stereo matching algorithm ,In the process of computing the disparity space, we obtain the disparity space based zero-mean normalized correlation ,and introduce the multi-window into the calculation and present the detailed way .In the process of obtaining the disparity ,we incorporation the continuous constraint to selecting the disparity where disparity isn't assigned. The experiment resulting on real world images shows the correctness of using the left-right line of sight detecting the occlusion, At the same time, we also learn that the result of multi-window are better than non-multi-window.Taking points by disparity consistency checking as GCPS, we obtain good disparity image through forcing scan-line optimize methods.Through researching the existing region growing method, we presented an improved algorithm unlike the existing region growing algorithms depending on the accuracy of the seeds .the algorithm can continue growing on the smoothing region and repetitive patterns region. Firstly, we obtain the initial set of seeds in the disparity space by matching the Harris points .then build the newly growing method in the disparity space. The results demonstrate it can trace out the disparity space. in the process of the growing ,it can automatically rectified the wrong matching, it guarantees matching accuracy and correctness in the smoothing and repetitive pattern region.