Defect Detection in Printed Circuit Boards with Pre-Trained Feature Extraction Methodology with Convolution Neural Networks

作     者：Mohammed A.Alghassab 

作者机构：Electrical Engineering DepartmentCollege of EngineeringShaqra UniversityRiyadh11911Saudi Arabia 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2022年第70卷第1期

页      面：637-652页

核心收录：

学科分类：0809[工学-电子科学与技术（可授工学、理学学位）] 08[工学] 

基　　金：The author would like to thank Deanship of Scientific Research at Shaqra University for their support to carry this work 

主　　题：Printed circuit board convolution neural network inception vgg16 data augmentation 

摘      要：Printed Circuit Boards(PCBs)are very important for proper functioning of any electronic *** are installed in almost all the electronic device and their functionality is dependent on the perfection of *** PCBs do not function properly then the whole electric machine might ***,keeping this in mind researchers are working in this field to develop error free *** these PCBs were examined by the human beings manually,but the human error did not give good results as sometime defected PCBs were categorized as ***,researchers and experts transformed this manual traditional examination to automated *** to this research image processing and computer vision came into actions where the computer vision experts applied image processing techniques to extract the ***,this also did not yield good ***,to further explore this area Machine Learning and Artificial Intelligence Techniques were *** this studywe have appliedDeep Neural Networks to detect the defects in the ***16and Inception networkswere applied to extract the relevant *** dataset was used in this study,it has 1500 pairs of both defected and non-defected *** pre-processing and data augmentation techniques were applied to increase the training *** neural networks were applied to classify the test *** results were compared with state-of-the art technique and it proved that the proposed methodology outperformed *** evaluation metrics were applied to evaluate the proposed *** 94.11%,Recall 89.23%,F-Measure 91.91%,and Accuracy 92.67%.