一种大场景下的棋盘靶标自动定位算法  

A kind of automatic chessboard target location algorithm used in a wide background environment

作　　者：林国余[1] 张为公[1] 

机构地区：[1]东南大学仪器科学与工程系,江苏南京210096

出　　处：《工业仪表与自动化装置》2008年第3期84-88,共5页Industrial Instrumentation & Automation

基　　金：江苏省交通科学研究计划资助项目(06C04)

摘　　要：在大场景且背景影响较大的测量环境下,目前常见的用于相机标定的角点提取算法无法准确获取棋盘靶标的角点,因此针对棋盘靶标图案,提出一种大场景下的棋盘靶标自动定位算法。该算法根据棋盘靶标的结构特征,首先由加权亮度矩和灰度梯度投影算法自动定位图像中可能的棋盘靶标区域,其次在棋盘子图像中利用Hough变换验证该棋盘靶标所在区域的有效性。实验结果表明,该算法能够在大场景中自动准确地定位出棋盘靶标区域,为后续进一步高精度角点定位奠定良好的基础。Since the common corner location algorithm used in camera calibration cannot work well in a wide background environment, this paper proposes a kind of automatic chessboard target location algorithm. According to the structural feature of the chessboard target, the weighted intensity moment and the gray gradient projection algorithm is used to get the possible target area, and then the Hough transformation is used in the chessboard sub - image to verify the located target area. Experiments show that the algorithm can locate the chessboard target sub - image automatically and correctly in the wide background environment,which provides a good foundation for further attaining the chessboard corner location up to a high accuracy.

关 键 词：棋盘靶标定位 亮度矩 HOUGH变换 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]