Robust Estimation of Trifocal Tensor Using Messy Genetic Algorithm  

作　　者：HUMingxing YUANBaozong TANGXiaofang 

机构地区：[1]InstituteofInformationScience,NorthernJiaotongUniversity,Beijing100044,China

出　　处：《Chinese Journal of Electronics》2003年第2期174-178,共5页电子学报（英文版）

摘　　要：Given three partially overlapping views of a scene from which a set of point or line correspondences have been extracted, 3D structure and camera motion pa-rameters can be represented by the trifocal tensor, which is the key to many problems of computer vision among three views. This paper addresses the problem of robust esti-mating the trifocal tensor employing a new method based on messy genetic algorithm, which uses each gene to stand for a triplet of correspondences, and takes every chromo-some as a minimum subset for trifocal tensor estimation.The method would eventually converge to a near optimal solution and is relatively unaffected by the outliers. Exper-iments with both synthetic data and real images show that our method is more robust and precise than other typical methods because it can efficiently detect and delete the bad corresponding points, which include both bad loca-tions and false matches.

关 键 词：三焦点张量 遗传算法 强估计 3D图像重建 

分 类 号：TN911.73[电子电信—通信与信息系统]