Research Of Parameters Calibration Method Of 3D Scan System

3D reconstruction is a newborn and promising technology and it is widely used in Reverse Engineering and rapid inspection of product etc. However, it is an important but difficult problem to propose a method which can be used to measure 3D surface of an object accurately and rapidly and recover the 3D shape. Camera calibration is a key step for 3D reconstruction from 2D images. The thesis is concentrated on two themes: location of calibration points and calibration algorithm.The precision of the data of calibration points is crucial to the calibration result. In this paper, a subpixel location algorithm based on the gray information of transition area in the edge of the object is presented. After getting the images of calibration points by CCD cameras, we can calculate the gradient of pixels, distill the pixels of transition area through a settled threshold, and use the position and gray information of these pixels to calculate the center of the calibration points. Meanwhile, the precision of different methods is compared based on the projection invariability of straight line.Based on the existing methods in the literature, the calibration method of cameras is further discussed in this thesis. After the whole course of 3D reconstruction is analyzed, system parameters are presented and the calibration mathematical model is put forward to calculate these parameters. In the end, the virtues and shortcomings of the methods are analyzed and the research direction is pointed out.