基于Harris度量的轮廓尖锐度CDA优化算法  

CDA optimization algorithm for contour sharpness based on the Harris metric

作　　者：于微波[1] 王国秀 李岩[1] 杨宏韬[1] 李昱 刘克平[1] YU Wei-bo;WANG Guo-xiu;LI Yan;YANG Hong-tao;LI Yu;LIU Ke-ping(School of Electrical and Electronic Engineering, Changchun University of Technology, Changchun 130012, China)

机构地区：[1]长春工业大学电气与电子工程学院,吉林长春130012

出　　处：《广西大学学报（自然科学版）》2021年第1期98-106,共9页Journal of Guangxi University（Natural Science Edition）

基　　金：吉林省发改委省级产业创新专项(2019C010);吉林省科技发展计划项目(20190303099SF)。

摘　　要：为了进一步提高工业现场中工件自动测量的精度,在详细分析了目前工业检测过程中角点提取准确度低、检测时间长等问题产生原因的基础上,提出了一种基于Harris度量的轮廓尖锐度工件图像角点检测优化算法。首先利用Freeman链码法提取工件图像的轮廓,通过计算轮廓尖锐度获取轮廓上的候选角点,然后针对两种不同类型的伪角点,分别采用Harris度量窗口和支撑区域自适应算法进行优化,最终得到工件图像的真实角点。将算法应用到了实际工件的尺寸测量中。相对于传统的角点检测算法,所提出的基于Harris度量的轮廓尖锐度CDA优化算法具有较好的准确性。In order to further improve the accuracy of automatic measurement of workpieces in the industrial field,an optimization algorithm for detecting sharpness of contour workpieces based on Harris metric was proposed on the basis of a detailed analysis of the causes of problems such as low accuracy of corner extraction and long detection time in current industrial inspection process.The algorithm first extracts the contour of the workpiece image by Freeman chain code method,obtains the candidate corner points on the contour by calculating the contour sharpness,then uses the Harris metric window and support region adaptive algorithm to optimize two different types of pseudo-corner points,and finally obtains the real corner points of the workpiece image.The algorithm was applied to the dimensional measurement of actual workpieces.Compared to traditional corner point detection algorithms,the proposed CDA optimization for contour sharpness based on the Harris metric has better alluracy.

关 键 词：角点检测 轮廓尖锐度 Harris度量 FREEMAN链码 支撑区域自适应 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]