摄像机标定中栅格型标定点的亚像素提取的研究  

Study on sub-pixel detection for grid target points in camera calibration

作　　者：陈大志[1] 王颖[1] 张广军[1] 

机构地区：[1]北京航空航天大学仪器科学与光电工程学院,北京100083

出　　处：《光学技术》2006年第1期53-55,58,共4页Optical Technique

基　　金：国家杰出青年科学基金资助项目(50125518)

摘　　要：介绍了一种新的摄像机标定用栅格型靶标标定点亚像素的提取方法。该方法先利用一个基于Hessian矩阵的形状算子判定栅格点的像素位置,然后对栅格点邻近区域内的图像灰度函数进行二阶泰勒展开,并通过计算该灰度曲面的局部极小值点来确定栅格中心点的亚像素位置。相对于传统的直线拟合求交点的方法,该算法计算过程简单,且具有较高的提取精度。A new sub-pixel detector for grid target points in camera calibration was proposed. The pixel position of grid point was detected by a new shape operator based on Hessian matrix. A second order Taylor polynomial describing the local intensity profile around the detected point was educed. The sub-pixel position of grid point was determined directly by calculating the local minimum point of this intensity profile. Compared with the traditional algorithm detecting the grid point by calculating the intersection of two fitted lines, this new algorithm yields simple calculation and accurate estimates.

关 键 词：摄像机标定 栅格型标定点 亚像素提取 

分 类 号：TN911.73[电子电信—通信与信息系统]