Stereo Matching Based On A Speckle-projected Stereovision System

To obtain three-dimensional (3D) information about the given surface has been increasing by importance for various purposes. Optical 3D sensing technology can be divided into active and passive approaches according to whether using structure light. The widely known passive scheme is stereovision, which is useful to measure surface with various environments. Stereo matching is one of the most important and difficult problem in stereovision technology. It has several problems to measure some special surfaces where there are not many reference points. Under such an environment, the abundance of reference points can produce matching mistakes. Thus an active system based on a structured light concept will be useful and a stereo system based on speckle projection has been proposed.It discusses in detail of the concept of the position random digital speckle, the principle of the stereovision and the matching method based on local correlation. A fast correlation matching method and a multi-window method are studied to improve the efficiency and accuracy respectively. Several typical surfaces are measured and the dense and accurate disparity maps have been acquired. The main result obtained can be summarized as following:The advantages of both passive and active optical sensing technologies are reviewed. A stereo system with digital speckle projection is presented. It isparticularly helpful for the reconstruction of the surface lack of reference points or texture.Correlation plays a key role in the stereo matching. The local correlation corresponding method is analyzed in detail. To improve the efficiency and accuracy of the matching, a fast correlation matching method and a multi-window method are studied.With the speckle projected stereo system, several typical surfaces abundance of reference points are computed. Accurate and dense disparity maps are achieved. The efficiency of our method is shown.