Deep learning-enhanced defects detection for printed circuit boards
Printed circuit boards (PCBs) are an important component of electronic devices. Therefore, ensuring the quality of such PCBs in the manufacturing process is crucial. Especially, cracks or scratches appearing on the PCB surface pose a significant hurdle, due to their minuscule size, making them the m...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123025001550 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583855170650112 |
|---|---|
| author | Van-Truong Nguyen Xuan-Thuc Kieu Duc-Tuan Chu Xiem HoangVan Phan Xuan Tan Tuyen Ngoc Le |
| author_facet | Van-Truong Nguyen Xuan-Thuc Kieu Duc-Tuan Chu Xiem HoangVan Phan Xuan Tan Tuyen Ngoc Le |
| author_sort | Van-Truong Nguyen |
| collection | DOAJ |
| description | Printed circuit boards (PCBs) are an important component of electronic devices. Therefore, ensuring the quality of such PCBs in the manufacturing process is crucial. Especially, cracks or scratches appearing on the PCB surface pose a significant hurdle, due to their minuscule size, making them the most challenging to address. In this work, we present a real-time automated algorithm for defects inspection of printed circuit boards (PCBs) in different lighting conditions. First, the Oriented FAST and Rotated BRIEF (ORB) algorithm extracts features from the input images, then the Brute-force matching method matches these features with the ORB features template. Next, the input images are calibrated to match the size and orientation of the template data by the RANSAC (Random Sample Consensus) algorithm. Finally, the defective areas on the PCB surface are segmented by using the U-NET (i.e., a type of convolutional neural network (CNN)) model. The proposed algorithm is tested in three different lighting conditions: low light, normal light, and high light conditions. Experimental studies are conducted on a representative PCB to evaluate the defect detection capacity of the proposed algorithm and the experimental results show that the proposed system works well in the three different lighting conditions with an accuracy of up to 97%, the detection speed is 12 frames per second (FPS). |
| format | Article |
| id | doaj-art-94ebed55bab94579969b920be1742ee5 |
| institution | Kabale University |
| issn | 2590-1230 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Engineering |
| spelling | doaj-art-94ebed55bab94579969b920be1742ee52025-01-28T04:14:50ZengElsevierResults in Engineering2590-12302025-03-0125104067Deep learning-enhanced defects detection for printed circuit boardsVan-Truong Nguyen0Xuan-Thuc Kieu1Duc-Tuan Chu2Xiem HoangVan3Phan Xuan Tan4Tuyen Ngoc Le5Faculty of Mechatronics, SMAE, Hanoi University of Industry, Hanoi, 11900, Viet NamFaculty of Electronic Engineering, Hanoi University of Industry, Hanoi, 11900, Viet Nam; Corresponding author.Faculty of Mechatronics, SMAE, Hanoi University of Industry, Hanoi, 11900, Viet NamFaculty of Electronics and Telecommunications, VNU-University of Engineering and Technology, Vietnam National University, Hanoi, 10000, Viet NamCollege of Engineering, Shibaura Institute of Technology, Tokyo, 135-8548, Japan; Principal corresponding author.Department of Electronic Engineering, Ming Chi University of Technology, Taipei 24301, Taiwan; Center for Reliability Engineering, Ming Chi University of Technology, Taipei 24301, TaiwanPrinted circuit boards (PCBs) are an important component of electronic devices. Therefore, ensuring the quality of such PCBs in the manufacturing process is crucial. Especially, cracks or scratches appearing on the PCB surface pose a significant hurdle, due to their minuscule size, making them the most challenging to address. In this work, we present a real-time automated algorithm for defects inspection of printed circuit boards (PCBs) in different lighting conditions. First, the Oriented FAST and Rotated BRIEF (ORB) algorithm extracts features from the input images, then the Brute-force matching method matches these features with the ORB features template. Next, the input images are calibrated to match the size and orientation of the template data by the RANSAC (Random Sample Consensus) algorithm. Finally, the defective areas on the PCB surface are segmented by using the U-NET (i.e., a type of convolutional neural network (CNN)) model. The proposed algorithm is tested in three different lighting conditions: low light, normal light, and high light conditions. Experimental studies are conducted on a representative PCB to evaluate the defect detection capacity of the proposed algorithm and the experimental results show that the proposed system works well in the three different lighting conditions with an accuracy of up to 97%, the detection speed is 12 frames per second (FPS).http://www.sciencedirect.com/science/article/pii/S2590123025001550Printed circuit boardU-NETDefect detectionScratches defectComputer vision |
| spellingShingle | Van-Truong Nguyen Xuan-Thuc Kieu Duc-Tuan Chu Xiem HoangVan Phan Xuan Tan Tuyen Ngoc Le Deep learning-enhanced defects detection for printed circuit boards Results in Engineering Printed circuit board U-NET Defect detection Scratches defect Computer vision |
| title | Deep learning-enhanced defects detection for printed circuit boards |
| title_full | Deep learning-enhanced defects detection for printed circuit boards |
| title_fullStr | Deep learning-enhanced defects detection for printed circuit boards |
| title_full_unstemmed | Deep learning-enhanced defects detection for printed circuit boards |
| title_short | Deep learning-enhanced defects detection for printed circuit boards |
| title_sort | deep learning enhanced defects detection for printed circuit boards |
| topic | Printed circuit board U-NET Defect detection Scratches defect Computer vision |
| url | http://www.sciencedirect.com/science/article/pii/S2590123025001550 |
| work_keys_str_mv | AT vantruongnguyen deeplearningenhanceddefectsdetectionforprintedcircuitboards AT xuanthuckieu deeplearningenhanceddefectsdetectionforprintedcircuitboards AT ductuanchu deeplearningenhanceddefectsdetectionforprintedcircuitboards AT xiemhoangvan deeplearningenhanceddefectsdetectionforprintedcircuitboards AT phanxuantan deeplearningenhanceddefectsdetectionforprintedcircuitboards AT tuyenngocle deeplearningenhanceddefectsdetectionforprintedcircuitboards |