摘要
为进一步提高摄像机参数标定的准确率,提出一种基于改进风驱动算法的摄像机标定优化方法。将摄像机内参作为空气微团的位置矢量,将标定结果的重投影误差作为空气微团的适应度值,通过迭代寻找适应度值最小的解。针对基本风驱动算法中固有参数的设置问题,通过引入协方差矩阵自适应进化策略自动确定。将该方法与张正友标定法、基于粒子群算法的标定法和基于基本风驱动算法的标定法作比较,实验结果表明,改进算法优化标定的摄像机内参精度更高,具有更快的收敛速度和更好的稳定性。
To further improve the accuracy of camera calibration,a camera calibration optimization method based on improved wind-driven algorithm was proposed.The internal parameters of the camera were taken as the position vector of the air micelles,the reprojection error of the calibration result was used as the fitness value of the air micelles,and the solution with the smallest fitness value was found through iteration.The problem of setting the inherent parameters in the basic wind-driven algorithm was automatically determined by introducing the covariance matrix adaptive evolution strategy.Compared with Zhang Zhengyou calibration method,particle swarm optimization based calibration method and basic wind-driven algorithm based calibration,experimental results show that the improved algorithm optimizes the camera’s internal parameters for higher accuracy,faster convergence,and better stability.
作者
任久斌
曹中清
REN Jiu-bin;CAO Zhong-qing(College of Mechanical Engineering,Southwest Jiaotong University,Chengdu 610031,China)
出处
《计算机工程与设计》
北大核心
2021年第4期942-948,共7页
Computer Engineering and Design
基金
国家自然科学基金项目(U1730131)。
关键词
摄像机标定
风驱动算法
协方差矩阵自适应进化策略
固有参数
重投影误差
camera calibration
wind-driven algorithm
covariance matrix adaptive evolution strategy
inherent parameters
reprojection error
