基于改进遗传算法的摄像机自标定方法  

Camera Self-calibration Method Based on Improved Genetic Algorithm

作　　者：杨亚男 贾渊[1] Yang Yanan;Jia Yuan(School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang 621010,China)

机构地区：[1]西南科技大学计算机科学与技术学院

出　　处：《计算机测量与控制》2020年第2期188-191,196,共5页Computer Measurement &Control

基　　金：四川省对地观测高分数据中心委托项目(19zh011201)

摘　　要：摄像机自标定技术不受标定板和相机运动轨迹的限制就能求解出摄像机的内参数矩阵,其标定过程简单、适用性强;由于传统的遗传算法在摄像机自标定参数优化过程中易出现过早收敛、停滞现象和解易陷入局部最优的问题,提出一种改进的遗传算法;首先,通过结合精英保留策略和随机联赛选择算法作为初始化种群的方法、改进轮盘赌选择方法、采用自适应杂交概率和变异概率方法对遗传算法进行改进;然后,将Hartley定义的简化Kruppa方程转化为目标函数,采用改进的遗传算法搜索目标函数的最优值;最后,实验结果表明,该方法能较好地缓解过早收敛和停滞显现,提高了精度。The camera self-calibration technology can solve the internal parameter matrix of the camera without being limited by the calibration plate and camera motion trajectory.The calibration process is simple and applicable.Because the traditional genetic algorithm is prone to premature convergence,stagnation and easy to fall into local optimum in the camera self-calibration parameter optimization process,an improved genetic algorithm is proposed.Firstly,the genetic algorithm is improved by combining the elite retention strategy and the random league selection algorithm as the method of initializing the population,improving the roulette selection method,adopting the adaptive hybridization probability and the mutation probability method;then,transforming the simplified Kruppa equation defined by Hartley for the objective function,the improved genetic algorithm is used to search for the optimal value of the objective function.Finally,the experimental results show that the method can better alleviate the premature convergence and stagnation and improve the accuracy.

关 键 词：摄像机自标定 遗传算法 KRUPPA方程 基础矩阵 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]