EDFW-YOLO: enhancing YOLOv8 for quantitative analysis of surface defects in hot-rolled strips

Abstract

Detecting surface defects on steel, especially in complex loading environments, poses significant challenges. In response, we introduce EDFW-YOLO, an algorithm built upon you only look once version 8 (YOLOv8) specifically designed for detecting surface defects on hot-rolled steel strips. Our method enhances multi-scale feature fusion through the incorporation of the multi-scale conversion module (C2f-EMSC). Additionally, we elevate detection accuracy by integrating the dynamic head target detection head, the focal modulation module, and the WIoU_Loss bounding box regression function. Experimental results on the NEU-DET dataset demonstrate that our optimized YOLOv8 model achieves the mean average precision (mAP) of 77.7%, with a 5.2% increase in network constraint rate. To adapt to different operating environments, it further improved the mAP to 78.5% through data enhancement. Verification results on PCB defect data show that the algorithm has excellent generalization ability. This optimized algorithm significantly improves the extraction and fusion of surface defect features on hot-rolled strip steel and serves as a valuable reference for surface defect detection in alloy materials.

Keywords

A

References

[1] Xu Y, Yang S, Zhang Xet al.. Steel surface defects detection based on improved faster R-CNN. 4th International Conference on Control and Robotics, December 2–4, 2022, Virtual, 2022, New York, IEEE353357[C]

[2] Qu Y, Wan B, Wang Cet al.. Optimization algorithm for steel surface defect detection based on PP-YOLOE. Electronics, 2023, 12(19): 4161 J]

[3] Pan K, Hu H, Gu P. WD-YOLO: a more accurate YOLO for defect detection in weld X-ray images. Sensors, 2023, 23218677 J]

[4] Guo Z, Wang C, Yang Get al.. MSFT-YOLO: improved YOLOv5 based on transformer for detecting defects of steel surface. Sensors, 2022, 2293467 J]

[5] Xie Y, Yin B, Han Xet al.. Improved YOLOv7-based steel surface defect detection algorithm. Mathematical biosciences and engineering, 2022, 211346-368 J]

[6] Yang B, Zhang L, He J. Research on rice disease recognition based on improved SPPFCSPC-G YOLOv5 network. PLOS one, 2023, 18(12): e0295661 J]

[7] Niu Z, Zhong G, Yu H. A review on the attention mechanism of deep learning. Neurocomputing, 2021, 452: 48-62 J]

[8] Redmon J, Farhadi A. YOLO9000: better, faster, stronger. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, July 21–26, 2017, Hawaii, USA, 2017, New York, IEEE7263-7271[C]

[9] REDMON J. YOLOv3: an incremental improvement[EB/OL]. (2018-04-08) [2024-12-22]. https://arxiv.org/abs/1804.02767.

[10] Yang G, Feng W, Jin J Tet al.. Face mask recognition system with YOLOv5 based on image recognition. 2020 IEEE 6th International Conference on Computer and Communications, December 11–14, 2020, Chengdu, China, 2020, New York, IEEE13981404[C]

[11] Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 17–24, 2023, Vancouver, BC, Canada, 2023, New York, IEEE74647475[C]

[12] Wang C Y, Bochkovskiy A, Liao H Y M. Scaled-YOLOv4: scaling cross stage partial network. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 20–25, 2021, Nashville, TN, USA, 2021, New York, IEEE1302913038[C]

[13] GE Z. YOLOx: exceeding YOLO series in 2021[EB/OL]. (2021-07-18) [2024-12-22]. https://arxiv.org/abs/2107.08430.

[14] Hussain M. YOLO-v1 to YOLO-v8, the rise of YOLO and its complementary nature toward digital manufacturing and industrial defect detection. Machines, 2023, 11(7): 677 J]

[15] Lou H, Duan X, Guo Jet al.. DC-YOLOv8: small-size object detection algorithm based on camera sensor. Electronics, 2023, 12102323 J]

[16] Lin W, Wu Z, Chen Jet al.. Scale-aware modulation meet transformer. Proceedings of the IEEE/CVF International Conference on Computer Vision, October 1–6, 2023, Paris, France, 2023, New York, IEEE6015-6026[C]

[17] Najibi M, Singh B, Davis L S. Autofocus: efficient multi-scale inference. Proceedings of the IEEE/CVF International Conference on Computer Vision, October 27–November 2, 2019, Seoul, Korea, 2019, New York, IEEE97459755[C]

[18] Nan Y, Ju J, Hua Qet al.. A-MobileNet: an approach of facial expression recognition. Alexandria engineering journal, 2022, 61(6): 4435-4444 J]

[19] Dai X, Chen Y, Xiao Bet al.. Dynamic head: unifying object detection heads with attentions. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 20–25, 2021, Nashville, TN, USA, 2021, New York, IEEE73737382[C]

[20] Yang J, Han F, Fan Het al.. Classification for dermoscopy images using convolutional neural networks based on region average pooling. IEEE access, 2018, 6: 65130-65138 J]

[21] Yang J, Li C, Dai Xet al.. Focal modulation networks. Advances in neural information processing systems, 2022, 35: 4203-4217[J]

[22] TONG Z, CHEN Y, XU Z, et al. Wise-IoU: bounding box regression loss with dynamic focusing mechanism[EB/OL]. (2023-01-24) [2024-12-22]. https://arxiv.org/abs/2301.10051.

[23] Song K, Yan Y. A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects. Applied surface science, 2013, 285: 858-864 J]

[24] Litjens G, Kooi T, Bejnordi B Eet al.. Deep learning for medical image analysis: a comprehensive review. Journal of medical image analysis, 2017, 42: 60-88 J]

[25] Liu S, Liu L, Tang Jet al.. Edge computing for autonomous driving: opportunities and challenges. Proceedings of the IEEE, 2019, 107(8): 1697-1716 J]

[26] HUANG W, WEI P. A PCB dataset for defects detection and classification[EB/OL]. (2019-01-24) [2024-12-22]. https://arxiv.org/abs/1901.08204.

RIGHTS & PERMISSIONS

Tianjin University of Technology