一种基于特征点的管材表面缺陷视觉检测方法  

A Visual Inspection Method for Pipe Surface Defect Based on Feature Point

作　　者：郑建聪 谢麒麟[1] 方挺 韩家明 董冲 ZHENG Jiancong;XIE Qilin;FANG Ting;HAN Jiaming;DONG Chong(Tube Pipe and Bar Business Unit,Baoshan Iron&Steel Co.,Ltd,Shanghai 201900;School of Electrical&Information Engineering,Anhui University of Technology,Maanshan 243032,China)

机构地区：[1]宝山钢铁股份有限公司钢管条钢事业部,上海201900 [2]安徽工业大学电气与信息工程学院,安徽马鞍山243032

出　　处：《安徽工业大学学报（自然科学版）》2022年第1期21-24,共4页Journal of Anhui University of Technology（Natural Science）

基　　金：国家自然科学基金项目(61971004)。

摘　　要：为快速准确检测管材表面缺陷,设计一种基于特征点的管材表面缺陷视觉检测方法。采集典型管材表面缺陷图像,构建图像样本集;指定图像感兴趣区域,减少干扰背景对缺陷检测的影响;采用ORB(oriented FAST and rotated BRIEF)算法检测图像中的点缺陷,使用FAST算子搜寻缺陷图像的特征点,将检测到的特征点设为圆心,以圆心与取点区域的形心连接线为横坐标构建特征点描述子;选取典型管材缺陷图像对所提方法进行仿真验证。结果表明,设计的检测方法检测准确率高、检测速度快,具备较高的工程实用价值,可为管材表面缺陷的自动检测提供预研基础。To detect the defects on the pipe surface accurately and timely,a visual inspection method for pipe surface defect based on feature point was proposed.An image sample set was constructed by collecting the surface defect images of typical pipe,and the influence of interference background on defect detection was reduced by specifying the region of interest of the image.The defect detection method based on ORB(oriented FAST and rotated BRIEF)feature point was used to detect the point defect in the image,and the FAST operator was used to search the feature point of the defect image,the detected feature points were set as the center of the circle,and the feature point descriptor was constructed with the connecting line between the center of the circle and the centroid of the point taking area as the abscissa.Typical pipe defect images were selected to verify the proposed method.The results show that the designed detection method has high detection accuracy,fast detection speed,high engineering practical value,and can provide a pre-research basis for automatic detection of pipe surface defects.

关 键 词：金属管材 表面缺陷 感兴趣区域 ORB特征点 视觉检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TG335.71[自动化与计算机技术—计算机科学与技术]