激光与光电子学进展, 2009, 46 (12): 99, 网络出版: 2009-12-17
一种基于OpenCV 的摄像机标定新方法 下载: 508次
A Novel Method of Camera Calibration Based on OpenCV
摄像机标定 改进型SUSAN 算子 镜头畸变 内部参数 camera calibration improved SUSANoperator OpenCV OpenCV lens distortion intrinsic parameter
摘要
针对基于计算机视觉技术的空间方位测量系统研制过程中摄像机标定问题,分析了通用摄像机模型,深入讨论了镜头的径向畸变和切向畸变对标定精度的影响以及相应的解决办法。介绍了2D 平面标定靶标制作规范,给出了一种改进型SUSAN 算子进行角点亚像素检测的方法,该方法具有速度快、定位精度高等优点。在VC++6.0 环境下调用OpenCV1.0 库函数实现了摄像机的标定,在实现过程中充分利用了OpenCV库函数,提高了标定精度和计算效率。
Abstract
For the camera calibration problem in research of spatial position measurement system based on computer vision technology,the general camera model is analysed,the influences and solutions of lens radial distortion and tangential distortion are discussed. A 2D planar pattern is built,and an improved SUSAN operator for sub -pixel corner detection is proposed,which has high speed and positioning accuracy. The camera calibration using the library functions of OpenCV1.0 in VC++6.0 is achieved,which improves the calibration accuracy and calculation efficiency.
李明金, 熊显名, 张绍兵. 一种基于OpenCV 的摄像机标定新方法[J]. 激光与光电子学进展, 2009, 46(12): 99. Li Mingjin, Xiong Xianming, Zhang Shaobing. A Novel Method of Camera Calibration Based on OpenCV[J]. Laser & Optoelectronics Progress, 2009, 46(12): 99.