基于机器视觉的PCB板电解电容极性检测  

PCB Electrolytic Capacitor Polarity Detection Based on Machine Vision

作　　者：包晓敏 王志豪 杨旭 BAO Xiao-min;WANG Zhi-hao;YANG Xu(School of Information Science and Technology,Zhejiang Sci-Tech University,Hangzhou 310018,China)

机构地区：[1]浙江理工大学信息学院,浙江杭州310018

出　　处：《测控技术》2020年第10期62-66,共5页Measurement & Control Technology

基　　金：国家自然科学基金青年科学基金(61801429)。

摘　　要：针对目前电解电容检测方法定位偏移较大、对噪声较为敏感,且极性判断准确性较差等问题,提出了一种基于轮廓特征与滑动窗口平滑度的电解电容检测方法,通过计算轮廓近似圆形状描述子与轮廓逼近折线角度的连续性特征,并结合广义霍夫变换圆检测来定位电解电容内圆。依据滑动窗口法计算电容圆环区域窗口平滑度,并对特征窗口累加计数来判断极性方向,最终实现电解电容检测。实验结果表明,该方法具有较高的准确率,能够适应电路板在线检测对时间的要求,并且对于电容形状不规则、有文字干扰以及噪声较多等情况具有良好的鲁棒性。An electrolytic capacitor detection method based on contour features and sliding window smoothness is proposed to solve the problems of large positioning deviation,sensitivity to noise and poor accuracy of polarity judgment. The inner circle of electrolytic capacitor was located by calculating the continuity features of the contour approximate circular shape descriptor and the angle of contour approaching polyline, and combining with the generalized Hough transform( GHT) circle detection. According to the sliding window method, the smoothness of the window in the capacitor ring area was calculated,and the characteristic windows were counted to determine the polarity direction. The electrolytic capacitor was detected. The experimental results show that this method has high accuracy,can meet the time requirements of online circuit board detection,and has good robustness for the case of irregular capacitance shape,text interference and more noise.

关 键 词：机器视觉 轮廓特征 形状描述子 平滑度 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] TP20[自动化与计算机技术—计算机科学与技术]