Applying Machine Learning to Construct a Printed Circuit Board Gold Finger Defect Detection System

Chien-Yi Huang,Pei-Xuan Tsai
DOI: https://doi.org/10.3390/electronics13061090
IF: 2.9
2024-03-16
Electronics
Abstract:Machine vision systems use industrial cameras' digital sensors to collect images and use computers for image pre-processing, analysis, and the measurements of various features to make decisions. With increasing capacity and quality demands in the electronic industry, incoming quality control (IQC) standards are becoming more and more stringent. The industry's incoming quality control is mainly based on manual sampling. Although it saves time and costs, the miss rate is still high. This study aimed to establish an automatic defect detection system that could quickly identify defects in the gold finger on printed circuit boards (PCBs) according to the manufacturer's standard. In the general training iteration process of deep learning, parameters required for image processing and deductive reasoning operations are automatically updated. In this study, we discussed and compared the object detection networks of the YOLOv3 (You Only Look Once, Version 3) and Faster Region-Based Convolutional Neural Network (Faster R-CNN) algorithms. The results showed that the defect classification detection model, established based on the YOLOv3 network architecture, could identify defects with an accuracy of 95%. Therefore, the IQC sampling inspection was changed to a full inspection, and the surface mount technology (SMT) full inspection station was canceled to reduce the need for inspection personnel.
engineering, electrical & electronic,computer science, information systems,physics, applied
What problem does this paper attempt to address?
### Problems the paper attempts to solve This paper aims to establish an automatic defect detection system to quickly identify the defects of gold fingers on printed circuit boards (PCBs). Specifically, the paper attempts to solve the following problems: 1. **Improve the efficiency of quality control**: - Traditional incoming quality control (IQC) mainly relies on manual sampling inspection. Although it saves time and cost, the false detection rate is still relatively high. - By introducing a machine - vision system, the paper hopes to reduce or eliminate the need for manual sampling inspection, thereby improving the detection efficiency and accuracy. 2. **Reduce the false detection rate**: - Manual sampling inspection has a relatively high false detection rate, which may lead to quality problems in the subsequent production process. - The automatic defect detection system can conduct a more comprehensive inspection of raw materials, reduce the false detection rate, and improve the overall production efficiency. 3. **Optimize image - processing techniques**: - During the training process of the deep - learning network, automatically update the parameters required for image - processing and inference operations to optimize the performance of defect detection. - By comparing the two algorithms, YOLOv3 and Faster R - CNN, select the most suitable model to establish the defect detection system. 4. **Realize full - inspection instead of sampling inspection**: - Currently, IQC is mainly based on manual sampling inspection, while the automatic defect detection system can achieve full - inspection, cancel the surface - mount technology (SMT) full - inspection station, and reduce the demand for inspectors. ### Summary The main objective of the paper is to establish an efficient automatic defect detection system by applying machine - learning techniques, especially deep - learning networks (such as YOLOv3), to improve the quality - control level of gold fingers on printed circuit boards (PCBs). This system aims to reduce the false detection rate, improve the detection efficiency, and ultimately realize the transition from sampling inspection to full - inspection.