Multi-Scale Lightweight Neural Network for Steel Surface Defect Detection
Source abstract
Abstract
Defect classification is an important aspect of steel surface defect detection. Traditional approaches for steel surface defect classification employ convolutional neural networks (CNNs) to improve accuracy, typically by increasing network depth and parameter count. However, this approach overlooks the significant memory overhead of large models, and the incremental gains in accuracy diminish as the number of parameters increases. To address these issues, a multi-scale lightweight neural network model (MM) is proposed. The MM model, with a fusion encoding module as its core, constructs a multi-scale neural network by utilizing the Gaussian difference pyramid. This approach enhances the network’s ability to capture patterns at different resolutions while achieving superior model accuracy and efficiency. Experimental results on a dataset from a hot-rolled strip steel plant demonstrate that the MM network achieves a classification accuracy of 98.06% in defect classification tasks. Compared to networks such as ResNet-50, ResNet-101, VGG, AlexNet, MobileNetV2, and MobileNetV3, the MM model not only reduces the number of model parameters and compresses model size but also achieves better classification accuracy.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Infrastructure Maintenance and Monitoring · Non-Destructive Testing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Intelligent Science & Engineering, Shenyang University, Shenyang 110044, China
School of Information Engineering, Shenyang University, Shenyang 110044, China
Funding
Shenyang Science and Technology Plan · Shenyang University
References (34)
Show source-provided references
1. He, Y., Wen, X., and Xu, J. (2022). A Semi-Supervised Inspection Approach of Textured Surface Defects under Limited Labeled Samples. Coatings, 12.10.3390/coatings12111707 ↗
2. An End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical Features10.1109/TIM.2019.2915404 ↗
3. Liu, Y., Yuan, Y., Balta, C., and Liu, J. (2020). A Light-Weight Deep-Learning Model with Multi-Scale Features for Steel Surface Defect Classification. Materials, 13.10.3390/ma13204629 ↗
4. Classification of Surface Defects on Steel Strip Images using Convolution Neural Network and Support Vector Machine10.1007/s11668-022-01344-6 ↗
5. Nizan, O., and Tal, A. (2023). k-NNN: Nearest Neighbors of Neighbors for Anomaly Detection. arXiv.
6. Shamsi, M., and Beheshti, S. (2023). Separability and Scatteredness (S&S) Ratio-Based Efficient SVM Regularization Parameter, Kernel, and Kernel Parameter Selection. arXiv.
7. Classification of hyperspectral remote sensing images with support vector machines10.1109/TGRS.2004.831865 ↗
8. Wen, X., Shan, J., He, Y., and Song, K. (2023). Steel Surface Defect Recognition: A Survey. Coatings, 13.10.3390/coatings13010017 ↗
9. Research progress on surface defect detection methods based on machine vision
10. Using Recurrent Neural Network Models for Early Detection of Heart Failure Onset10.1093/jamia/ocw112 ↗
11. Simonyan, K., and Zisserman, A. (2015). Very Deep Convolutional Networks for Large-Scale Image Recognition. arXiv.
12. He, K., Zhang, X., Ren, S., and Sun, J. (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. Available online: https://openaccess.thecvf.com/content_cvpr_2016/html/He_Deep_Residual_Learning_CVPR_2016_paper.html.10.1109/CVPR.2016.90 ↗
13. Detecting Welding Defects in Steel Plates using Machine Learning and Computer Vision Algorithms
14. Automated Steel Surface Defect Detection and Classification Using a New Deep Learning-Based Approach10.1007/s00521-022-08112-5 ↗
15. Image Processing-Based Pitting Corrosion Detection Using Metaheuristic Optimized Multilevel Image Thresholding and Machine-Learning Approaches10.1155/2020/6765274 ↗
16. A New Steel Defect Detection Algorithm Based on Deep Learning10.1155/2021/5592878 ↗
17. Steel-surface defect detection using a switching-lighting scheme10.1364/AO.55.000047 ↗
18. Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., and Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv.
19. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.-C. (2018, January 18–23). MobileNetV2: Inverted Residuals and Linear Bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA. Available online: https://openaccess.thecvf.com/content_cvpr_2018/html/Sandler_MobileNetV2_Inverted_Residuals_CVPR_2018_paper.html.10.1109/CVPR.2018.00474 ↗
20. Zhang, X., Zhou, X., Lin, M., and Sun, J. (2018, January 18–23). ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA. Available online: https://openaccess.thecvf.com/content_cvpr_2018/html/Zhang_ShuffleNet_An_Extremely_CVPR_2018_paper.html.10.1109/CVPR.2018.00716 ↗
21. Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., and Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50× fewer parameters and <0.5 MB model size. arXiv.
22. Choi, S., and Choi, J. (2023). Arithmetic Intensity Balancing Convolution for Hardware-aware Efficient Block Design. arXiv.
23. Hu, J., Shen, L., and Sun, G. (2018, January 18–23). Squeeze-and-Excitation Networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA. Available online: https://openaccess.thecvf.com/content_cvpr_2018/html/Hu_Squeeze-and-Excitation_Networks_CVPR_2018_paper.html.10.1109/CVPR.2018.00745 ↗
24. Wan, C., Ma, S., and Song, K. (2022). TSSTNet: A Two-Stream Swin Transformer Network for Salient Object Detection of No-Service Rail Surface Defects. Coatings, 12.10.3390/coatings12111730 ↗
25. Bergstrom, A.C., Conran, D., and Messinger, D.W. (2023). Gaussian Blur and Relative Edge Response. arXiv.
26. Scale-Space Theory: A Basic Tool for Analyzing Structures at Different Scales10.1080/757582976 ↗
27. Self-grouping convolutional neural networks10.1016/j.neunet.2020.09.015 ↗
28. Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.-Y., and Berg, A.C. (2015). SSD: Single Shot MultiBox Detector. arXiv, Available online: https://arxiv.org/abs/1512.02325v5.
29. Lowe, D.G. (1999, January 20–27). Object recognition from local scale-invariant features. Proceedings of the Seventh IEEE International Conference on Computer Vision, Corfu, Greece.10.1109/ICCV.1999.790410 ↗
30. Simoncelli, E.P., and Freeman, W.T. (1995, January 23–26). The steerable pyramid: A flexible architecture for multi-scale derivative computation. Proceedings of the International Conference on Image Processing, Washington, DC, USA.
31. Edge Detection of an Image Based on Extended Difference of Gaussian10.11648/j.ajcst.20190203.11 ↗
32. Dropout: A Simple Way to Prevent Neural Networks from Overfitting
33. Zhang, C., Hu, H., Fang, D., and Duan, J. (2020, January 11–13). The CCD sensor video acquisition system based on FPGA&MCU. Proceedings of the 2020 IEEE 9th Joint International Information Technology and Artificial Intelligence Conference (ITAIC), Chongqing, China.10.1109/ITAIC49862.2020.9339037 ↗
34. Loshchilov, I., and Hutter, F. (2019). Decoupled Weight Decay Regularization. arXiv.