基于灰度值的异形构件激光切割缺陷检测  

Detection of defects in laser cutting of irregular components based on grayscale value

作　　者：张天成[1] 张春艳[2] ZHANG Tiancheng;ZHANG Chunyan(Anhui Technical College of Mechanical and Electrical Engineering,Wuhu 241000 China;Bengbu University,Bengbu 233030 China)

机构地区：[1]安徽机电职业技术学院,安徽芜湖241000 [2]蚌埠学院,安徽蚌埠233030

出　　处：《新余学院学报》2024年第5期56-64,共9页Journal of Xinyu University

基　　金：安徽省2020年质量工程项目校企合作实践教育基地“安徽机电职业技术学院芜湖马尔克斯智能科技有限公司实践教育基地”(2020sjjd030)。

摘　　要：基于机器视觉构建了异形构件成像系统,通过摄像机模型获取异形构件切割图像,采用小波变换增强采集的图像后,计算图像的灰度值,从而增强缺陷区域与周围背景的灰度差异,分割缺陷区域,提高缺陷检测的有效性。完成缺陷区域的分割后,提取图像的纹理特征并输入到支持向量机中,完成异形构件激光切割缺陷分类检测。测试结果显示:该方法图像增强效果较好,可恢复图像的整体清晰度;对比度、相关性均在0.924以上;能量结果均在0.12以下,能精准完成异形构件的表面缺陷的分类识别。An imaging system for irregular components based on machine vision is established.Cutting images of irregular components are obtained through the camera model of the system.After wavelet transform is used to enhance the collected images,the grayscale value of the images is calculated,thereby enhancing the grayscale difference between the defect area and the surrounding background,segmenting the defect area and improving the effectiveness of defect detection.After the segmentation of the defect area is completed,the texture features of the image are extracted and input into the support vector machine to complete the laser cutting defect classification detection of irregular components.The test results show that this method has a good image enhancement effect and can restore the overall clarity of the image;The contrast and correlation are both above 0.924;The energy results are all below 0.12,accurately classifying and identifying surface defects of irregular components.

关 键 词：灰度值 异形构件 激光切割 缺陷检测 成像系统 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]