基于灰度特征的棋盘格内角点检测算法  

Checkerboard grid interior corner point detection algorithm based on grayscale features

作　　者：翟凯玥 李静[1] 鲁济帅 崔露露 ZHAI Kaiyue;LI Jing;LU Jishuai;CUI Lulu(College of Weapon Science and Technology,Xi’an Technological University,Xi’an 710021,China)

机构地区：[1]西安工业大学兵器科学与技术学院,西安710021

出　　处：《激光杂志》2023年第12期40-46,共7页Laser Journal

基　　金：国防基础科研项目(No.JCKY2016606B001);西安市科技计划项目(No.2019220514SYS020CG042)。

摘　　要：标定物特征点的定位精度能直接影响相机标定的精准度,为提高检测的精度和效率,提出了一种基于灰度特征的棋盘格内角点检测的改进算法。通过分析角点的灰度分布特征,首先利用3×3环形邻域模板对初始角点预筛选;随后通过计算初始角点的BW算子响应值实现对棋盘格边缘角点及伪角点的剔除;最后结合改进的角点响应函数确定最终内角点。实验结果表明:运用改进算法在不同光照及噪声的影响下检测结果无漏检及误检,单幅图像所需检测时间为0.225 s,平均重投影误差结果为0.045像素,能为相机的高精度标定提供更精准可靠的数据。The accuracy of the positioning of the feature points of the object can directly influence the accuracy of the camera calibration.In order to improve the accuracy and efficiency of detection,this paper proposes an improved algorithm for detecting the corner points inside the checkerboard grid based on grayscale features.By analyzing the grayscale distribution characteristics of corner points,the initial corner points are first pre-screened using a 3×3 circular neighborhood template.Subsequently,the rejection of corner points and pseudo corner points at the edge of the checkerboard grid is achieved by calculating the response values of the BW operator for the initial corner points.Finally,the final inner corner points are determined by combining the improved corner point response function.The experimental results show that the improved algorithm of this paper has no missed detection and false detection under the influence of different illumination and noise.The detection time for a single image is 0.225 s,and the average reprojection error result is 0.045 pixels,which can provide more accurate and reliable data for high precision calibration of the camera.

关 键 词：角点检测 棋盘格 灰度特征 BW算子 相机标定 

分 类 号：TN206[电子电信—物理电子学]