硬度图像角点检测算法  

The Hardness Testing Based on a Novel Corner Detection Algorithm

作　　者：曲福恒[1] 杨勇[1] 胡雅婷[2] 陈学广[1] 

机构地区：[1]长春理工大学计算机科学技术学院,长春130022 [2]吉林农业大学信息技术学院,长春130118

出　　处：《长春理工大学学报（自然科学版）》2011年第3期135-138,共4页Journal of Changchun University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金(10926157);国家十一五科技支撑计划(2009BAE69B01)

摘　　要：硬度检测图像中存在大量杂质与噪声,目标区域的边缘模糊且形状复杂,传统角点检测算法无法检测出目标区域的角点,也很难应用直线拟合或直线检测算法实现角点的精确定位。针对硬度检测中面临的复杂环境下角点检测任务,提出一种新的基于旋转灰度变化的角点检测算法。与传统的角点检测算法相比,该方法对噪声和杂质具有更好的鲁棒性,能够在背景复杂、边缘模糊的情况下准确检测出目标区域的角点位置,具有更好的适用性。在硬度图像上与传角点检测算法及直线求交确定角点方法的对比实验表明了算法的有效性。The conventional corner detection algorithm can not exactly find the real corners of the object region of the hardness testing image,because there are much noise and impurities in the image and the edge of its object region is fuzzy and bent with a complicated shape.It is also difficult to extract the corner position by the line fitting or detection algorithm because of the complicated shape of its edge.In order to solve such corner detection problem of hardness im-age with complex background,a novel corner detection algorithm based on the gray difference of the rotation is pro-posed in this paper.Compared with the conventional algorithms,the proposed algorithm is more robust to the noises and impurities,and can detect the corners of the image in the complicated background.The contrast experimental re-sults show the effectiveness of the algorithm.

关 键 词：角点检测 硬度图像 图像处理 边缘模糊 鲁棒性 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]