平行多基线立体视觉图像校正

Rectification of images for parallel multiple-baseline stereo vision

介绍了现有双目立体视觉校正算法的原理和过程,分析了其应用于多基线立体视觉系统的局限性,得出其产生原因是各摄像机光心位置不共线造成图像平面不共面.保持三目立体视觉系统中两个摄像机的光心位置不变,利用光心和基线构建三角形,将第三个摄像机的光心投影到前两个摄像机光心连线构成的基线上,使各摄像机的光心共线,然后使用双目视觉校正算法对三个摄像机进行校正.结果表明,算法可正确有效地校正平行三目立体视觉图像对,并可推广至摄像机数目多于三个的平行多基线立体视觉系统.

The principle and process of current rectification algorithm for binocular stereo vision images were discussed. After analysis of the above algorithm's limitation that cannot be applied in the rectification of multiple-baseline stereo vision images, the causation that the noncollinear principal point of each camera resulting in non-coplanarity was obtained. Keeping two of them unchanged in parallel trinocular stereo vision system, a triangle was established, which was made of principal points and baselines. The third principal point was projected to the baseline made of the former two, so each was colinear. Then three cameras were rectified by binocular stereo vision images rectification algorithm. The experimental results show that this algorithm can not only rectify the images for parallel trinocular stereo vision system, but also be extended to the parallel multiple-baseline stereo vision system made of more than three cameras.