Research On 3D Surface Recovery Technique Based On 2D Gray Image

The technique which 3D surface shape is recovered from 2D gray image (shape from shading, SFS) is one of the key techniques of 3D shape recovery in computer field. Due to its feature, which can restore 3D shape information just from one gray-level image, recently, SFS gets great improved and develops rapidly in methodology and application and become a new research of 3D passive reconstruction.At the beginning of this thesis, a comprehensive survey about SFS research work is given, and the recent research focuses in this field are summarized. For consideration of principle of photometry, the illumination of three local illumination models that are Lambert, Phong and Oren-Nayar model, are determined. And with SFS analytical principle, the correct illumination model for SFS is fixed already. On the base of understanding for SFS problem in nature, a systematic summary is made for all current SFS algorithms at present, which are minimization approaches, propagation approaches, local approaches, and linear approaches. Meanwhile, based on principle of photometry and theory of numerical solution for nonlinear partial differential equation, the corresponding algorithm based on the Lambert reflectance model for minimization approach is provided.Meanwhile, an algorithm of reverse calculation in the spot gray of virtual space by given height and gradient has also been presented, which is in order to provide each pixel a real color and get an image with the strongest feeling of reality. reality. By this way, the complexity and error because of the camera calibration can be avoided and a perfect research flat will be acquired.At least, with the result of experiment, which shows that the algorithm is feasible and has stated application value, a better 3D reconstruction result has been gotten.