Printed circuit board inspection using computer vision

Abstract

The inspection of electronic components, especially printed circuit boards (PCBs), has greatly benefited from the advancements in computer vision technology. With the miniaturization of electronic components, defects on PCBs are now often found in smaller or micro-sized forms. This poses a significant challenge for automated optical inspection methods to effectively detect and identify such small objects. The primary objective of this study is to address the issue of fault detection in printed circuit boards (PCBs). To achieve this, the study employs various image processing techniques to carry out the inspection process. These image processing operations play a crucial role in preparing the images for defect analysis. Once the image processing operations are completed, the study proceeds to classify the identified defects in the segmented regions using a support vector machine (SVM) classifier. The SVM classifier is trained to categorize the defects based on the extracted features and their respective class labels. This classification step plays a critical role in accurately identifying and characterizing the detected defects. To evaluate the effectiveness of this study, a comparison is made with earlier works in the field. This allows for a comprehensive assessment of the proposed methodology and its performance in comparison to existing approaches. By benchmarking against previous works, the study provides valuable insights into the advancements and improvements achieved in PCB defect detection.