立体视觉测量中的图像匹配策略研究  

Research of image matching strategy on stereo vision measurement

作　　者：陈棣湘[1] 罗飞路[1] 潘孟春[1] 

机构地区：[1]国防科技大学机电工程与自动化学院,长沙410073

出　　处：《光学技术》2002年第5期392-394,共3页Optical Technique

摘　　要：立体视觉测量广泛应用于机器人导向和各种柔性制造系统中 ,其中图像特征的准确匹配始终是一个难题。推导了共极性线方程 ,采取相关匹配和特征匹配相结合的方法来提高图像匹配的速度和精度 ,利用建模估计得到匹配像点的精确坐标 ,使三维测量精度达到目标距离的 1/10 0 0Stereo vision measurement has being used in robot guidance and various flexible manufacturing systems, but the correct matching of image features remains a difficult problem. The epipolar line equation is deduced and the method of interrelated matching incorporated with feature matching to improve the speed and precision of image matching is applied. At the same time a model estimation method is used to find the exact coordinates of matching image points. The relative measurement precision of three dimensions achieves to around 1/10000 of the distance between the camera to the measured object.

关 键 词：立体视觉测量 图像匹配 机器人 FMS 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置] TP391.41[自动化与计算机技术—控制科学与工程]