The Research Of Moving Vehicle 3D Reconstruction Based On Binocular Stereo Vision

The 3-D reconstruction technology of computer vision is a significant field of computer vision, and what's more, it can help realizing 3-D untouched measurement and fast reconstruction. This technology is a main way of 3-D reconstruction in the future because it is efficient and automatic.In this paper, the object of research is the Moving Vehicle. The research on 3D geometric modeling of moving vehicle and its implement through computer vision 3D reconstruction technique and machine technique in any background solve the problem of low precision, poor reliability and stake test recognition. Thus, stereovision system calibration, feature extraction and stereo matching of moving vehicle are forced in this paper.Calibration: To get the precise camera parameters, the linear and non-linear methods were combined. First, Direct Linear Transform model was used to get the projection matrix. Then the matrix was decomposed based on the restraint condition, and the intrinsic parameters of camera were estimated. These estimates were fed to the non-linear optimizer.Feature Extraction: A improved HIS image difference model based on sequences image difference and second image difference is presented. A method of self-adaptive division and elimination of the object shadow to extract the moving objects without background knowledge is adopted. And the target regions corner and edge detectionStereo Match: A stereo matching approach based on edge feature is proposed. The matched corners were used to guide the edge matching. The procedure of the correspondence is fulfilled through the edge-detection, the similarity check and continuous relaxation iteration algorithm.Finally, the results of 3-D reconstruction are given. The simulations are performed by MATLAB language.