摘要
为能准确提取物体空间信息实现图像三维重建,对摄像机透视畸变模型及透视投影模型进行研究,在张正友标定法的基础上,提出基于共面点的摄像机非线性标定方法。利用棋盘格角点的三维坐标和相应的图像坐标运算获取畸变参数,并通过非线性最小二乘法对其进行优化,从而得到精确的摄像机标定参数。实验结果表明,改进的摄像机标定方法相比张正友标定法具有更高的标定精度和标度速度,适用于单摄像机的标定。
In order to accurately extract the spatial information of objects to realize 3D reconstruction of the image,the camera lens distortion model and the lens projection model of the camera are studied. Based on the calibration method proposed by Zhang Zhengyou, an improved nonlinear calibration method for cameras based on coplanar points is presented. The distortion parameters are obtained by using the 3D coordinates of the chessboard corners and the corresponding image coordinates. They are optimized by the nonlinear least square method, thus obtaining the accurate camera calibration parameters. Experimental results show that the improved camera calibration method has higher calibration accuracy and calibration speed compared with the Zhang Zhengyou calibration method, which is suitable for the calibration of single camera.
出处
《计算机工程》
CAS
CSCD
北大核心
2016年第8期289-293,共5页
Computer Engineering
基金
江苏省产学研联合创新基金资助项目"冷链物联网关键技术研发与应用"(BY2014023-36)
关键词
计算机视觉
三维重建
摄像机标定
透镜畸变
共面点
棋盘格角点
computer vision
3D reconstruction
camera calibration
lens distortion
coplanar point
chessboard corner
