Article Details

Abstract

The stable operation of power supply boards is crucial for various electronic devices. At present, the appearance inspection of power supply boards still mostly relies on manual visual inspection. This paper proposes an application method of YOLOv13 in the detection of missing components on power supply boards. Firstly, the training method of the YOLOv13 model is introduced, and the method for selecting image augmentation parameters is analyzed. By comparing the training results of models trained with different parameters, the optimal YOLOv13n model is obtained. Secondly, the trained model is verified and can accurately identify the marked components. Finally, a missing-component detection system for power supply boards is designed to obtain the camera video stream, invoke the trained YOLOv13n model for inference and prediction, and compare the component coordinates with those of a standard power supply board, thereby determining whether there are missing components on the tested power supply board and providing specific information about the missing components. Experimental results show that the missing-component detection system for power supply boards designed in this paper can effectively detect whether there are missing components on power supply boards.

Keywords

power supply board; missing component detection; image recognition; YOLOv13

References

1. [1] Zhang Liangjing, Wu Chunjiang, Wu Liuqiang, etc Application research of improved YOLOv13 in detection and recognition of sugarcane red rot disease [J]. Wireless Internet Technology, 2026, 23 (4): 1-10
2. [2] Yang Nianxiang, Cao Lijia, Yang Yang, etc A review of research on PCB defect detection algorithms based on machine vision [J]. Computer Science and Exploration, 2025, 19 (4): 901-915
3. [3] Liu Kaipei, Li Boqiang, Qin Liang, etc A review of the application of deep learning object detection algorithms in insulator defect detection of overhead transmission lines [J]. High Voltage Technology, 2023, 49 (9): 3584-3595
4. [4] Cheng Jinfeng, Fang Guisheng, Gao Huifang Research progress on machine vision technology for surface defect detection [J]. Computer Application Research, 2023, 40 (4): 967-977
5. [5] Yang Zeqing, Zhang Mingxuan, Chen Yingshu, etc Research progress on surface defect detection methods based on machine vision [J]. Modern Manufacturing Engineering, 2023 (4): 143-156
6. [6] Luo Zhen, Jiang Yu, Li Weicheng Road defect detection model based on improved YOLO11n [J]. Microelectronics and Computer, 2025, 42 (11): 25-36
7. [7] Liu Chuanyang, Wu Yiquan Research progress on visual inspection methods for transmission lines based on deep learning [J]. Chinese Journal of Electrical Engineering, 2023, 43 (19): 7423-7446
8. [8] Geng Chuang, Song Dejie, Cao Lijia Research progress of YOLO algorithm in object detection [J]. Journal of Weapon Equipment Engineering, 2022, 43 (9): 162-173
9. [9] Lei M Q，Li S Q，Wu Y H，et al. YOLOv13: Real-Time Object Detection with Hypergraph-Enhanced Adaptive Visual Perception [PP/OL].arXiv:2506.17733 [cs.CV].https://doi.org/10.48550/arXiv.2506.17733.
10. [10] Redmon J，Divvala S，Girshick R，et al. You Only Look Once: Unified, Real-Time Object Detection [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)，2016：779-788.

About the authors

Wang Kai (1989-), male, Han ethnicity, from Huai'an, Jiangsu, teaching assistant, master's student, research interests include image processing, deep learning and YOLO algorithms.