基于离散2-D小波多级分解的电容器外观缺陷视觉检测方法  

Visual Inspection Method of Capacitor Appearance Defects Based on Discrete 2-D Wavelet Multi-Level Decomposition

作　　者：张剑[1] 高云 何栋 ZHANG Jian;GAO Yun;HE Dong(Artificial Intelligence Department,Shanxi Polytechnic College,Taiyuan Shanxi 030006,China;School of Computer and Network Engineering,Shanxi Datong University,Datong Shanxi 037009,China;Intelligent Control Department,Shanxi Railway Vocational and Technical College,Taiyuan Shanxi 030013,China)

机构地区：[1]山西职业技术学院人工智能系,山西太原030006 [2]山西大同大学计算机与网络工程学院,山西大同037009 [3]山西铁道职业技术学院智能控制系,山西太原030013

出　　处：《电子器件》2024年第5期1255-1260,共6页Chinese Journal of Electron Devices

基　　金：山西省教育科学“十四五”规划课题项目(GH-220795)。

摘　　要：电容器外观破损、凸起等缺陷直接影响器件的生产质量。目前电容器微小外观缺陷检测难度较大,导致视觉检测效率较低。为此提出了基于离散2-D小波多层分解的电容器外观缺陷视觉检测方法。采用同态滤波处理去除光照对电容器外观视觉检测结果的影响。利用像素点灰度值确定图像边缘点位置,提取电容器外观缺陷区域。应用离散2-D小波分解的方法对其展开多级分解。再差分统计电容器图像的外观缺陷纹理特征。将特征输入Mahalanobis分类器中,完成电容器外观缺陷视觉检测。仿真结果表明,所提方法可以较好检测电容的各种缺陷,召回率最低值是94.9%,误检率最高值为9.8%,漏检率均在10%以内,电容器缺陷检测效果较好。The defects of capacitor such as breakage and bulge affect the quality of the device.It is very difficult to detect the small appearance defects of capacitors,which leads to low efficiency of visual inspection.A visual inspection method of capacitor appearance defects based on discrete 2-D wavelet multilayer decomposition is proposed.Homomorphic filtering is used to remove the influence of illumination on the visual inspection results of capacitors.Pixel gray value is used to determine the position of image edge points to extract the appearance defect region.Discrete 2-D wavelet decomposition is used to unfold the multilevel decomposition.Appearance defect texture characteristics of the capacitor image are differentially counted.The feature is input into Mahalanobis classifier to complete the visual defect detection.Simulation results show that the proposed method can detect all kinds of defects of capacitors,the minimum recall rate is 94.9%,the maximum false detection rate is 9.8%,and the missing detection rate is less than 10%.

关 键 词：多尺度特征 同态滤波处理 2-D小波分解 Mahalanobis分类器 电容缺陷 视觉检测 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]