Abstract
Steel may have several surface faults, like scratches, pitted surfaces, and so forth, which can severely affect the steel’s quality and, as a result, hurt an enterprise’s ability to make a profit. These defects are a result of unavoidable elements connected to material qualities and steel production technology. In contrast to traditional object detection, surface defect detection must detect small defects and some defects with excessive aspect ratios. Fine features and the great positional precision of the surface defects make the identification tasks exceedingly difficult. The deep learning network of larger size leads to poor real-time performance in detecting the flaws of the steel surface. In this study, the original YOLOV5 algorithm was enhanced to increase the precision and effectiveness of fault detection on steel surfaces. First, to give more attention to small defects, squeeze-and-excite (SELayers) are embedded in different channels. Second, to recognize small defects on the surface, a small target recognition layer (STRLayers) was introduced. Third, to perfect the excessive aspect ratio of detection, complete intersection over union (CIoU) loss function is used. The Northeastern University detection dataset on steel surfaces was taken for the proposed model evaluation. The precision and effectiveness of the proposed SESC YOLO were increased by 11% in detecting the flaws on the steel surface.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Su F, Wang S (2022) Improving the algorithm study of YOLO in steel surface defect detection. Int J Mater 9:26–34
Xi J, Shentu L, Hu J, Li M (2017) Automated surface inspection for steel products using computer vision approach. Appl Opt 56(2):184–192
He Y, Song K, Meng Q, Yan Y (2019) An end-to-end steel surface defect detection approach via fusing multiple hierarchical features. IEEE Trans Instrum Measur 69(4):1493–1504
Qing YAO, Jin F, Jian T, Xu W, Zhu X, Yang B, Jun LU et al (2020) Development of an automatic monitoring system for rice light-trap pests based on machine vision. J Integr Agricult 19(10):2500–2513
Gyimah NK, Girma A, Mahmoud MN, Nateghi S, Homaifar A, Opoku D (2021) A robust completed local binary pattern (RCLBP) for surface defect detection. In: Proceedings of the 2021 IEEE international conference on systems, man, and cybernetics (SMC), pp 1927–1934
Jeon Y-J, Choi D, Yun JP, Kim SW (2015) Detection of periodic defects using dual-light switching lighting method on the surface of thick plates. ISIJ Int 55(9):1942–1949
Suvdaa B, Ahn J, Ko J (2012) Steel surface defects detection and classification using SIFT and voting strategy. Int J Softw Eng Appl 6(2):161–166
Song K, Yan Y (2013) A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects. Appl Surf Sci 285:858–864
Ren S, He K, Girshick R, Sun J (2015) Faster r-cnn: towards real-time object detection with region proposal networks. Adv Neural Inform Process Syst 28:70
Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 779–788
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C-Y, Berg AC (2016) Ssd: single shot multibox detector. European conference on computer vision. Springer, Cham, pp 21–37
Hatab M, Malekmohamadi H, Amira A (2020) Surface defect detection using YOLO network. In: Proceedings of SAI intelligent systems conference. Springer, Cham, pp 505–515
Li M, Wang H, Wan Z (2022) Surface defect detection of steel strips based on improved YOLOv4. Comput Electr Eng 102:108208
Ning Z, Mi Z (2021) Research on surface defect detection algorithm of strip steel based on improved YOLOV3. J Phys Conf Ser 1907(1):012015
Kou X, Liu S, Cheng K, Qian Y (2021) Development of a YOLO-V3-based model for detecting defects on steel strip surface. Measurement 182:109454
Zeqiang S, Bingcai C (2022) Improved Yolov5 algorithm for surface defect detection of strip steel. Artificial intelligence in China. Springer, Singapore, pp 448–456
Shi J, Yang J, Zhang Y (2022) Research on steel surface defect detection based on YOLOv5 with attention mechanism. Electronics 11(22):3735
Yeung C-C, Lam K-M (2022) Efficient fused-attention model for steel surface defect detection. IEEE Trans Instrum Measur 71:1. https://doi.org/10.1109/TIM.2022.3176239
Wang M, Yang W, Wang L, Chen D, Wei F, Liao Y (2023) FE-YOLOv5: feature enhancement network based on YOLOv5 for small object detection. J Vis Commun Image Represent 90:103752. https://doi.org/10.1016/j.jvcir.2023.103752
Wan D, Lu R, Wang S, Shen S, Xu T, Lang X (2023) YOLO-HR: improved YOLOv5 for object detection in high-resolution optical remote sensing images. Remote Sens 15:614. https://doi.org/10.3390/rs15030614
Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 7132–7141
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Kavitha, S., Baskaran, K.R., Santhiya, K. (2023). SESC-YOLO: Enhanced YOLOV5 for Detecting Defects on Steel Surface. In: Shukla, P.K., Mittal, H., Engelbrecht, A. (eds) Computer Vision and Robotics. CVR 2023. Algorithms for Intelligent Systems. Springer, Singapore. https://doi.org/10.1007/978-981-99-4577-1_17
Download citation
DOI: https://doi.org/10.1007/978-981-99-4577-1_17
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4576-4
Online ISBN: 978-981-99-4577-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)