TY - GEN
T1 - MS-YOLOv5s
T2 - 20th International Conference on Intelligent Computing, ICIC 2024
AU - Wang, Chenchen
AU - Zhou, Mian
AU - Liang, Yuan
AU - Pan, Weiwei
AU - Gao, Zan
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.
PY - 2024
Y1 - 2024
N2 - The detection of surface defects in steel is crucial for maintaining high product quality standards and preventing financial losses due to inferior goods. However, the subtle imperfections that escape human observation present a significant challenge to existing algorithms. In response to this pressing need, our research introduces an advanced approach, MS-YOLOv5s, designed to better distinguish between background and defects. Firstly, our study presents a new neck module, the comprehensive position feature pyramid networks (CPFPN), which improves the precision of detecting barely noticeable flaws. This is achieved by using the spatial channel attention module (SCAM) on intermediate feature maps to retain more positional information from the original image. Moreover, this innovative method adopts multi-scale learning, dynamically adjusting the input image size during training to amplify the differences between defects and background. MS-YOLOv5s achieves 80.5% and 65.7% mean average precision (mAP) respectively on the NEU-DET and GC10-DET datasets, demonstrating robust performance across various scenarios and outperforming many methods in identifying defects on the steel surface.
AB - The detection of surface defects in steel is crucial for maintaining high product quality standards and preventing financial losses due to inferior goods. However, the subtle imperfections that escape human observation present a significant challenge to existing algorithms. In response to this pressing need, our research introduces an advanced approach, MS-YOLOv5s, designed to better distinguish between background and defects. Firstly, our study presents a new neck module, the comprehensive position feature pyramid networks (CPFPN), which improves the precision of detecting barely noticeable flaws. This is achieved by using the spatial channel attention module (SCAM) on intermediate feature maps to retain more positional information from the original image. Moreover, this innovative method adopts multi-scale learning, dynamically adjusting the input image size during training to amplify the differences between defects and background. MS-YOLOv5s achieves 80.5% and 65.7% mean average precision (mAP) respectively on the NEU-DET and GC10-DET datasets, demonstrating robust performance across various scenarios and outperforming many methods in identifying defects on the steel surface.
KW - Comprehensive position feature pyramid networks
KW - Multi-scale learning
KW - Spatial channel attention module
KW - Steel surfaces
UR - http://www.scopus.com/inward/record.url?scp=85201092066&partnerID=8YFLogxK
U2 - 10.1007/978-981-97-5609-4_31
DO - 10.1007/978-981-97-5609-4_31
M3 - Conference Proceeding
AN - SCOPUS:85201092066
SN - 9789819756087
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 404
EP - 415
BT - Advanced Intelligent Computing Technology and Applications - 20th International Conference, ICIC 2024, Proceedings
A2 - Huang, De-Shuang
A2 - Zhang, Chuanlei
A2 - Guo, Jiayang
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 5 August 2024 through 8 August 2024
ER -