摘要
针对工业生产中不同检测环境下轴承滚子自动检测时存在的误检问题,提出一种基于Hough变换圆检测和边缘模板匹配的轴承滚子检测方法。首先,使用同态滤波算法对图像光照进行校正,以增强图像的对比度,同时采用Hough变换圆检测算法获取轴承滚子的轮廓,并根据检测结果在原始图像中截取感兴趣区域;然后,利用边缘模板匹配算法对轴承滚子模板和感兴趣区域进行匹配筛选,获得已装配轴承滚子的全部定位点,再进一步基于定位点进行圆拟合,从而找到缺失滚子所在位置;最后,通过实验可知所提方法的正确检测率比传统预测匹配差阈值化方法和高斯加权均值分割法分别提高29%和37%,验证了所提方法的有效性。
Aiming at the problem of misdetection in automatic detection of bearing rollers under different inspection environments in industrial production,a bearing roller detection method based on Hough transform circle detection and edge template matching was proposed.First,the image illumination is corrected by using a homomorphic filtering algorithm to enhance the contrast of the image.At the same time,the contour of the bearing roller is obtained by using the Hough transform circle detection algorithm,and the region of interest is intercepted from the original image based on the detection results;on this basis,the edge template matching algorithm is used to match and filter the bearing roller template and the region of interest to obtain all the positioning points of the assembled bearing rollers,and then perform a circle fitting based on the positioning points to find the location of the missing roller;Finally,experiments show that the correct detection rate of the proposed method is 29%and 37%higher than that of the traditional prediction matching threshold method and Gaussian weighted mean segmentation method,respectively.The effectiveness of the proposed method is verified.
作者
段志达
魏利胜
刘小珲
丁坤
DUAN Zhida;WEI Lisheng;LIU Xiaohui;DING Kun(School of Electrical Engineering,Anhui Polytechnic University,Wuhu 241000,China;Shanghai Oushuo Intelligent Packaging Technology Co.,Ltd,Shanghai 201416,China)
出处
《安徽工程大学学报》
CAS
2020年第4期60-68,共9页
Journal of Anhui Polytechnic University
基金
安徽省自然科学基金资助项目(1908085MF215)
安徽工程大学中青年拔尖人才基金资助项目(2016BJRC008)。
关键词
同态滤波
HOUGH变换
模板匹配
轴承滚子检测
homomorphic filtering
hough transform
template matching
bearing roller inspection
