摘要
两步法是一种典型的多阶段最优化摄像机标定方法,其优点是标定精度高,容易实现。该算法的关键是第一步计算出来的外部参数和与畸变无关的内部参数的初值。由于初值的计算不是纯粹的线性问题,计算起来比较复杂。本文对此进行了改进,采用一种新的方法计算初值。另外,由于图像坐标值的精度直接影响标定精度,本文为此采用了数学形态学的膨胀和腐蚀方法,有效地克服了噪声的影响。实验结果表明,与平面模板两步法进行比较,本文算法减少了运算量,标定的结果适用于精度较高的场合。
The Two-Step Method is a classical multistage optimization algorithm. It is easy to implement and can get accurate results.The most important part of the algorithm is the initial value of external parameters and distortion-free internal parameters through the first step.The computation is complicated because it is not a pure linear problem.In this paper, we improve the computation procedure and adopt a new method to get the initial value.In addition,the precision of image coordinates affects the accuracy of camera calibration directly,so we deal with the images using the mathematical morphology and eliminate the noise infection effectively.The experimental results show that the algorithm in this paper decreases the operating quantities compared with the Two-step Method, and the calibration results are applicable on the occasions which require high precision.
出处
《计算机工程与科学》
CSCD
2007年第1期52-55,共4页
Computer Engineering & Science
基金
国家自然科学基金资助项目(60573079)
湖南省自然科学基金资助项目(03jjy6025)
湖南省教育厅科研资助项目(03C227)
关键词
两步法
摄像机标定
内部参数
外部参数
two-step method
camera calibration
internal parameters
external parameters
