The application of camera calibration methods based on structured-light measurement

Abstract

There are two steps in using 3D computer vision to extract metric information from 2D images. Camera calibration is the first step, and this paper analyses and compares the three popular calibration methods: typical linear calibration, Tsai's two-stage calibration and Zhang's planar pattern calibration by measuring the position of space points. It is concluded that Zhang's method has the best practicability and validity, and this has been verified by experiment. In the structured light vision system, the accuracy of center position of the stripe is of great importance for that of the whole measuring system. Processing the image of stripe is the second important step. We adopted an improved arithmetic based on Harris to extract the points of the planar pattern. At last, using it to obtain a cylinder's radius and prove it simple and available.