Research Of 3D Reconstruction Methods Based On Computer Vision

With the development of computer, microelectronics and communication, 3D shape recovery and modeling are the first step of the measurement chain in a number of applications, ranging from the reverse engineering process, to CAD, to a number of machine vision and virtual reality application. However, it is an important but difficult problem to device a method which can be used to measure 3D surface of an object accurately and rapidly and recover the 3D shape.In this paper, a method named Grating Projection is presented which captures the surface topography of a 3D object without any contact. The deformed grating pattern appears when the common rectangular grating or sine grating is projected onto the object surface, which includes information about the surface topography of 3D object. The deformed stripes input into a computer by CCD camera are processed by a series of procedure and then is analyzed according to the mathematic model.The calibration method is discussed mainly because camera calibration is a key step of 3D shape recovery. After the whole course of 3D reconstruction is analyzed, system parameters are presented and the calibration mathematical model is put forward to calculate this parameters. In the end, the virtues and shortcomings of the method are analyzed and the further research direction is pointed out.