基于SAPSO算法的单目视觉标定与测量

Monocular Vision Calibration and Measurement Based on SAPSO Algorithm

针对双目立体视觉测量存在视场小、特征点匹配困难的缺陷,分析摄像机非线性模型的畸变影响,利用模拟退火粒子群优化(SAPSO)算法在解空间的搜索能力,建立摄像机标定适应度函数实现精确标定,借助激光测距实现单目立体测量。实验结果表明,该方法具有较好的标定结果和较高的测量精度。

Aiming at the problem that stereoscopic vision measurement exists the lack of visual field and feature point is difficult to match,this paper analyzes the non-linear model of camera distortion effects,and uses Simulated Annealing Particle Swarm Optimization（SAPSO） algorithm to get the solution of camera calibration model＇s over determined equation.It uses laser ranging to achieve the realization of monocular vision measurement.Experimental results demonstrate the better calibration results and measurement precision of this algorithm.