Research and Analysis of Automobile Wheel Hub Surface Defect Detection Algorithm based on YOLOv3-spp
ZHANG Zhen-yu,LIU Yang,LIU Fu-cai
1. Engineering Research Center of the Ministry of Education for Intelligent Control System and Intelligent Equipment, Yanshan University, Qinhuangdao, Hebei 066004, China
2. CITIC Dicastal Co.Ltd, Qinhuangdao,Hebei 066004, China
3. Hebei High-end Equipment Industry Technology Research Institute, Qinhuangdao, Hebei 066004, China
Abstract According to the requirements of surface defect detection environment and detection index in the production process of automobile hub, aiming at the problems of low efficiency and low accuracy of traditional manual detection methods, an automatic defect detection method based on YOLOv3-spp (You Only Look Once v3-Spatial Pyramid Pooling Network) network is proposed. In this method, firstly, the defect area is extracted by image slicing, and then the extracted defect image is enhanced to form a data set to train the YOLOv3-spp deep learning network, what's more the hub defects are detected and compared by using different neural networks and data set screening methods. The experimental results show that: On the dataset collected from the industrial site, the trained YOLOv3-spp neural network can accurately locate and identify four types of defects: point-like, linear, oily sludge and pinhole, with an average accuracy of 84.5%, 93.4%, 95.4% and 89.5% respectively. The detection speed of a single image can reach 35ms, meeting the real-time requirements of detection. Furthermore, the detection accuracy is better than two common neural networks: Faster R-CNN(Faster Region-Convolutional Neural Networks) and SSD(Single Shot MultiBox Detector).
Received: 04 August 2022
Published: 21 September 2023
ZHANG Zhen-yu,LIU Yang,LIU Fu-cai. Research and Analysis of Automobile Wheel Hub Surface Defect Detection Algorithm based on YOLOv3-spp[J]. Acta Metrologica Sinica, 2023, 44(9): 1375-1382.
Li F, Tang Y J, Yuan W Q. The research on surface defects detection of motorcycle hub［J］. Machinery Design & Manufacture, 2020(8): 296-299.
Wang Z W, Guo B, Hu X F, et al. Internal Groove Defect Detection Method of Brake Master Cylinder Based on FCOS Neural Network［J］. Acta Metrologica Sinica, 2021, 42(9): 1225-1231.
Song H, Li Z. Design of visual inspection system for automobile hub surface defect based on industrial robot［J］. Computer Measurement and Control, 2018, 26(9): 13-16, 22.
Jiao T Y, Wang M Q, Zhang J S, et al. Hub X-ray image enhancement based on wavelet analysis and pseudo color processing［J］. Automation & Instrumentation, 2020, 35(1): 47-51.
Redmon J, Farhadi A. Yolov3: An Incremental Improvement［C］//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, USA, 2018.
Han K, Sun M, Zhou X, et al. A new Method in Wheel Hub Surface Defect Detection: Object Detection Algorithm Based on Deep Learning［C］//IEEE. International Conference on Advanced Mechatronic Systems (ICAMechS). Xiamen, China, 2017: 335-338.
Wang C, Zhang X F, Liu C, et al. Detection method of wheel hub weld defects based on the improved YOLOv3［J］. Optics and Precision Engineering, 2021, 29(8): 1942-1954.
Guo R Q, Wang M Q, Zhang J S, et al. Hub Defect Segmentation Based on U-Net Convolutional Neural Network［J］. Automation & Instrumentation, 2020, 35(4): 43-47.
Zhang Y Y, Ren T F, Wen Y T. A Method for Detecting Internal Defects of Metal Lattice Structure Based on YOLOv3 Algorithm［J］. Acta Metrologica Sinica, 2022, 43(1): 7-13.
Zhao N N, Tao Y, Li F, et al. Visual saliency superpixel image detection method for surface defect of high-speed rail hub surface defect of high-speed rail hub［J］. Science Technology and Engineering, 2019, 19(32): 230-235.
Zhu C P, Yang Y B. Online Detection Algorithm of Automobile Wheel Surface Defects Based on Improved Faster-RCNN Model［J］. Surface Technology, 2020, 49(6): 359-365.
［14］
Yu Y, Wang M, Wang Z, et al. Surface Defect Detection of Hight-speed Railway Hub Based on Improved YOLOv3 Algorithm［C］//2021 IEEE 4th Advanced Information Management, Communicates, Electronic and Automation Control Conference (IMCEC). Chongqing, China, 2021: 1386-1390.
［20］
Ren S, He K, Girshick R, et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks［C］//Advances in Neural Information Processing Systems (NIPS). Montreal, Canada, 2015.
Liu W, Anguelov D, Erhan D, et al. SSD: Single shot multibox detector［C］//In European Conference on Computer Vision (ECCV). Amsterdam, Netherlands, 2016.
Guo B S, Wu W W, Fu Q, et al. Color Difference Detection of Polycrystalline Silicon Cells Based on Support Vector Machine Classification Strategy［J］. Acta Metrologica Sinica, 2019, 40(6): 1013-1019.
Zhao H W, Zhao Y C, Qi X Y, et al. Research on surface defect inspection algorithms of automobile hub based on deep learning［J］. Modular Machine Tool & Automatic Manufacturing Technique, 2019(11): 112-115.
［16］
Tang C, Feng X, Wen H, et al. Semantic Segmentation Network for Surface Defect Detection of Automobile Wheel Hub Fusing High-Resolution Feature and Multi-Scale Feature［J］. Applied Sciences. 2021, 11(22): 10508.
［18］
Zhang P, Zhong Y, Li X. SlimYOLOv3: Narrower, Faster and Better for Real-Time UAV Applications［C］//2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW). Seoul, Korea, 2019: 37-45.
2023, Vol. 44 
