Multi-position Based Space Position Setting System

Multi-position based space position setting system essentially is a 3D reconstruction system which is based on the theory of binocular stereo vision. Binocular stereo vision is an important branch of three-dimensional vision. It can simulate human vision to measure the three-dimensional information of the Objects in a variety of conditions. Research on binocular stereo vision is significant for visual physiology and engineering application.A binocular stereo vision technology system mainly consists of several parts: Video Capture, camera calibration, Feature Extraction, 3-D matching and 3-D reconstruction. This article focuses on some of the above parts: First of all, the article summarized and summed up the research status and research methods of the binocular stereo vision technology at home and abroad. Secondly, in order to be used in the broadcast video packaging, the method for camera calibration required to be simple and flexible. Through studying and analyzing several typical methods of camera calibration, Zhang's flexible new camera calibration method is chosen and used in our system. Thirdly, stereo matching is a key technique in stereo vision. Based on a through study of available bibliography, stereo matching is studied and discussed in details. In order to be suitable for the application, the Birchfield algorithm is chosen and used in this system. The system gets the disparity map though the stereo matching. Finally, according to the basic principles of stereo vision and the calculation method of the three-dimensional coordinates, the system completed the calculation of the three-dimensional coordinates. And the system obtain the information of point set in specific region.At last, a system of camera calibration and 3D reconstruction is implemented in this paper, which is based on the function library of OpenCV.