一种新型棋盘格角点检测算法

摘要

Aiming at the corner detection for calibration board of chessboard pattern in the camera, a kind of cor-ner detection algorithm has been put forward based on the image pixel distribution. Firstly, the even zone points are removed by estimating the number of similar points around the pixel point 8 . Then the upper matrix of diago-nal is extracted according to the feature of height symmetry in chessboard pattern image to determine whether the central point is the corner candidate. At the end, through the calculation of the selected corner candidates getting the coordinate in average and the final angle point is obtained. The experimental result shows that the method can solve the testing problem of the chessboard angle points in a better way.%针对摄像机标定中棋盘格标定板的角点检测，给出了一种基于图像像素点分布的角点检测算法。首先通过判断像素点8邻域范围内相似点个数除去平坦区域点，然后根据棋盘格图像高度对称的特性提取对角线上矩阵，通过此矩阵来判定中心点是否为候选角点，最后通过计算候选角点的算术平均求取最终角点坐标。