一种线结构光视觉传感器标定方法

在线结构光视觉传感器的标定中,结构光光平面上三维世界标定点的获取一直是一个难题。主要是因为靶 标上已知的三维世界点很难恰好位于结构光光平面上。针对这个问题,提出了一种基于双重交比不变的线结构光 视觉传感器的标定方法,设计了相应的标定靶标。利用双重交比不变性,可以获取光平面上任意数量的高精度的 三维标定点,从而解决了线结构光视觉传感器三维标定点不易获取且数量少的问题,提高了线结构光视觉传感器 的标定精度。试验结果表明该方法可以满足大型工件的结构光三维视觉检测精度的要求。

Calibration approach for structured-lighted-stripe vision sensor

The problem associated with calibrating a structured-light-stripe vision sensor is that known world points on the calibration target do not normally fall onto the light stripe plane illuminated from the projector. To address this problem, a method basing on the invariance of double cross ratio is proposed for calibrating structured-light-stripe vision sensor, and the calibration target is designed. By this method, arbitrary amount of 3D world calibration points on the structured light plane can be easily obtained, thus the problem that it is difficult to obtain adequate 3D world calibration points is settled and the calibration accuracy of structured-light-stripe vision sensor is definitely improved. The experiments conducted suggest that the proposed calibration method is of high accuracy and is applicable to many structured light based 3D visual inspection tasks about large workpieces.