A Two-Stage Industrial Defect Detection Framework Based on Improved-YOLOv5 and Optimized-Inception-ResnetV2 Models
Source abstract
Abstract
Aiming to address the currently low accuracy of domestic industrial defect detection, this paper proposes a Two-Stage Industrial Defect Detection Framework based on Improved-YOLOv5 and Optimized-Inception-ResnetV2, which completes positioning and classification tasks through two specific models. In order to make the first-stage recognition more effective at locating insignificant small defects with high similarity on the steel surface, we improve YOLOv5 from the backbone network, the feature scales of the feature fusion layer, and the multiscale detection layer. In order to enable second-stage recognition to better extract defect features and achieve accurate classification, we embed the convolutional block attention module (CBAM) attention mechanism module into the Inception-ResnetV2 model, then optimize the network architecture and loss function of the accurate model. Based on the Pascal Visual Object Classes 2007 (VOC2007) dataset, the public dataset NEU-DET, and the optimized dataset Enriched-NEU-DET, we conducted multiple sets of comparative experiments on the Improved-YOLOv5 and Inception-ResnetV2. The testing results show that the improvement is obvious. In order to verify the superiority and adaptability of the two-stage framework, we first test based on the Enriched-NEU-DET dataset, and further use AUBO-i5 robot, Intel RealSense D435 camera, and other industrial steel equipment to build actual industrial scenes. In experiments, a two-stage framework achieves the best performance of 83.3% mean average precision (mAP), evaluated on the Enriched-NEU-DET dataset, and 91.0% on our built industrial defect environment.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Advanced Neural Network Applications · Domain Adaptation and Few-Shot Learning
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Center for Robotics, School of Control Science and Engineering, Shandong University, Jinan 250061, China
Engineering Research Center of Intelligent Unmanned System, Ministry of Education, Jinan 250061, China
College of Electronic Information and Optical Engineering, Nankai University, Tianjin 300350, China
Sinotruk Industry Park Zhangqiu, Sinotruk Jinan Power Co., Ltd., Jinan 250220, China
References (31)
Show source-provided references
1. Knowledge of ultra-high-strength steel used in aerospace and other industries
2. Ultrasound image defect classification based on genetic algorithm optimized support vector machine
3. A random forest classification algorithm for polarized SAR images with comprehensive multi-features
4. A defect-detection method of split pins in the catenary fastening devices of highspeed railway based on deep learning10.1109/TIM.2020.3006324 ↗
5. Machine vision and nondestructive inspection of steel plate surface defects
6. Surface defect automatic detection algorithm based on principal component analysis
7. Xiao-Cong, L. (2014, January 15–16). A hybrid SVM-QPSO model based ceramic tube surface defect detection algorithm. Proceedings of the 2014 Fifth International Conference on Intelligent Systems Design and Engineering Applications, Hunan, China.10.1109/ISDEA.2014.15 ↗
8. Ferguson, M., Ak, R., Lee, Y.-T.-T., and Law, K.H. (2017, January 11–14). Automatic localization of casting defects with convolutional neural networks. Proceedings of the 2017 IEEE International Conference on Big Data (Big Data), Boston, MA, USA.10.1109/BigData.2017.8258115 ↗
9. Railway subgrade defect automatic recognition method based on improved faster R-CNN
10. Improved YOLOV3 target recognition algorithm based on adaptive eged optimization
11. (2021, March 01). GitHub.YOLOV5-Master. Available online: https://github.com//ultralytics//yolov5.git//.
12. Wu, C., Wen, W., Afzal, T., Zhang, Y., and Chen, Y. (2017, January 21–26). A compact DNN: Approaching GoogLeNet-Level accuracy of classification and domain adaptation. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.88 ↗
13. Kim, D., Park, S., Kang, D., and Paik, J. (2019, January 8–11). Improved center and scale prediction-based pedestrian detection using convolutional block. Proceedings of the 2019 IEEE 9th International Conference on Consumer Electronics (ICCE-Berlin), Berlin, Germany.10.1109/ICCE-Berlin47944.2019.8966154 ↗
14. Spatial pyramid pooling in deep convolutional networks for visual recognition10.1109/TPAMI.2015.2389824 ↗
15. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., and Berg, A.C. (2016, January 11–14). SSD: Single shot multibox detector. Proceedings of the European Conference on Computer Vision, Amsterdam, The Netherlands.10.1007/978-3-319-46448-0_2 ↗
16. Wang, W., Xie, E., Song, X., Zang, Y., Wang, W., Lu, T., Yu, G., and Shen, C. (November, January 27). Efficient and accurate arbitrary-shaped text detection with pixel aggregation network. Proceedings of the IEEE/CVF International Conference on Computer Vision, Seoul, Korea.
17. Zeiler, M.D., Taylor, G.W., and Fergus, R. (2011, January 6–13). Adaptive deconvolutional networks for mid and high level feature learning. Proceedings of the 2011 International Conference on Computer Vision, Barcelona, Spain.10.1109/ICCV.2011.6126474 ↗
18. Wang, Q., Wu, B., Zhu, P., Li, P., Zuo, W., and Hu, Q. (2020, January 14). ECA-Net: Efficient channel attention for deep convolutional neural networks. Proceedings of the Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.10.1109/CVPR42600.2020.01155 ↗
19. Szegedy, C., and Vanhoucke, V. (2016, January 12–17). Inception-v4, Inception-ResNet and the impact of residual connections on learning. Proceedings of the Thirty-first AAAI Conference on Artificial Intelligence, Phoenix, AZ, USA.10.1609/aaai.v31i1.11231 ↗
20. He, K., Zhang, X., and Ren, S. (2016, January 27–30). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, NV, USA.10.1109/CVPR.2016.90 ↗
21. Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., and Wojna, Z. (2016, January 21–26). Rethinking the inception architecture for computer vision. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, HI, USA.10.1109/CVPR.2016.308 ↗
22. Woo, S., Park, J., Lee, J.Y., and Kweon, I.S. (2018, January 8–14). Cbam: Convolutional block attention module. Proceedings of the European conference on computer vision (ECCV), Munich, Germany.10.1007/978-3-030-01234-2_1 ↗
23. Split-Attention Multiframe Alignment Network for ImageRest oration10.1109/ACCESS.2020.2967028 ↗
24. A tutorial on the cross-entropy method10.1007/s10479-005-5724-z ↗
25. The pascal visual object classes (voc) challenge10.1007/s11263-009-0275-4 ↗
26. Bochkovskiy, A., Wang, C.-Y., and Liao, H.-Y.M. (2020). YOLOv4: Optimal speed and accuracy of object detection. arXiv.
27. Multi-Scale Target Detection Algorithm Based on Attention Mechanism
28. Deep Learning for Computer Vision with Python Practitioner Bundle
29. Simon, M., Rodner, E., and Denzler, J. (2016). ImageNet pre-trained models with batch normalization. arXiv.
30. Akiba, T., Suzuki, S., and Fukuda, K. (2017). Extremely large minibatch SGD: Training ResNet-50 on ImageNet in 15 minutes. arXiv.
31. Menegola, A., Tavares, J., Fornaciali, M., Li, L.T., Avila, S., and Valle, E. (2017). RECOD titans at ISIC challenge 2017. arXiv.