一种结合遗传算法和LM算法的摄像机自标定方法  

A Camera Self-calibration Method Based on Genetic Algorithm and LM Algorithm

作　　者：范晶晶[1] 刘峰[1,2] 徐琼[1] 

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210003 [2]南京邮电大学江苏省图像处理与图像通信重点实验室,江苏南京210003

出　　处：《南京邮电大学学报（自然科学版）》2011年第5期23-26,32,共5页Journal of Nanjing University of Posts and Telecommunications：Natural Science Edition

基　　金：国家自然科学基金(61071166;61001152;61071091;JHB10-13)资助项目

摘　　要：提出了一种遗传算法和LM算法相结合的摄像机自标定方法。首先利用遗传算法获得若干组摄相机内外参数的初始值,接着利用LM算法对每一组参数进行优化,然后选取映射误差最小的那组参数。如果这组参数的映射误差满足要求,则将这组参数作为摄相机标定的最终结果;否则利用遗传算法对刚才得到的若干组参数进行交叉和变异操作,再利用LM算法优化每一组参数,直至映射误差满足要求。最后将摄像机标定结果用于视差估计,以判断标定结果的正确性。实验结果表明,提出的结合LM算法和遗传算法的摄像机自标定方法是正确有效的。A camera self-calibration method is proposed by combining genetic algorithm with LM algorithm.Firstly,we use genetic algorithm to get several groups of initial value of camera parameter,then we use the LM algorithm to optimize each group of parameter,and select the group of parameter with minimum projection error.If the minimum projection error meets the requirement,this group of parameter is the final result of camera self-calibration.Otherwise,we use genetic algorithm to cross and mutate these groups of parameters,and reuse LM algorithm to optimize each group of parameter until the projection error meets the requirement.Finally,we estimate the disparity with the camera parameters from camera self-calibration to judge the correctness of these parameters.Experimental results show that the proposed camera self-calibration method based on LM algorithm and genetic algorithm is correct and effective.

关 键 词：摄像机自标定 遗传算法 LM算法 视差估计 

分 类 号：TN919.81[电子电信—通信与信息系统]