The Research Of 3D Reconstruction Method Based On The Theory Of Computer Vision

With the development of computer technology, the research to computer vision field becomes a popular problem gradually. Computer vision is expansion and stretching of human, its quilt more and more broadly applies to each aspect. 3D reconstruction is one important and classical problem of computer vision field , is also the final step to imitate eye visual performance, is the 3D information restoring of the object. The paper is based on the theory of stereo vision , study and discuss respectively to the 3D reconstruction which is based on the theory of binocular stereo matching and the 3D information restoring which is based on the theory of SFS.In the first method which is based on binocular stereo matching , this paper used Zhang' s method to calibrate the camera, then detected and extracted the feature points by the Harris's method and accomplished the stereo matching with the gray scale correlation method, finally computed the 3D points of the target, and got the 3D model by Delaunay triangulation and texture mapping. In the second method which is based on SFS, this paper designed the different experiment condition accroding to the different target, then did the experiment and built the model or gray scale-height, compute the height of the target, acomplished the 3D imformation restoring. Two kinds method have got comparatively satisfied result equally.