When transporting goods at ports, the long-term use of containers can lead to various defects, and timely detection of defects is crucial. To address the issues of defects being highly integrated with the background and having minimal color differences in container defect detection, we propose a Distinguish-Background-YOLOv8 (DB-YOLOv8) algorithm. Adding the CA attention mechanism to the Backbone network of YOLOv8 improves the ability to recognize and locate targets. The experimental results show that the model has a mean average precision (mAP) of 83.4%, which is 2.4% higher than YOLOv8. Compared with current mainstream algorithms, it has better performance in container defect detection, and its detection speed can meet the practical requirements. It has high application prospects and value in container defect detection.

Yongzhao Liu, Fengguang Jia, Limin Yu, Xingkun Li, Yuan Liu, Guodong Chen, Guangrui Wu, Yujie Yin. Research on Container Defect Detection Based on DB-YOLOv8. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 11: 135-141. https://doi.org/10.25236/AJCIS.2024.071118.

[1] Zhu X. Market and Development Trends of Container Logistics Industry[J]. Marine Equipment/Materials & Marketing, 2022, 30(01):78-80.DOI:10.19727/j.cnki.cbwzysc.2022.01.025.

[2] Chen G. The current situation and trend of container transportation development in Chinese ports[J]. China Shipping Gazette, 2024, (30):45-47.

[3] Ma L, Zhu C, Zhou R. Container damage detection method based on YoLov4 algorithm[J]. Journal of Shanghai Maritime University, 2021, 42(4):114-118. DOI:10.13340/j.jsmu.2021.04.018.

[4] Guo S, Zhang H, Ge X, et al. Intelligent Container Defect Detection System Based on Video AI Algorithm[J]. Science & Technology of Ports, 2022, (06):21-22+42.

[5] Wang Z, Gao J, Zeng Q, et al. Multitype damage detection of container using CNN based on transfer learning[J]. Mathematical Problems in Engineering, 2021, 2021(1): 5395494.

[6] İmamoğlu Z E, Tuğlular T, Baştanlar Y. Container damage detection and classification using container images[C]//2020 28th Signal Processing and Communications Applications Conference (SIU). IEEE, 2020: 1-4. 

[7] Xiao J, Mei H, Yao H. Research on the Method of Container Incomplete Inspection Based on Edge Detection[J]. Port Operation, 2022, (05):39-43.

[8] Bahrami Z, Zhang R, Wang T, et al. An end-to-end framework for shipping container corrosion defect inspection[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-14.

[9] Pei X, Liu J, Bai X, et al. Damage detection algorithm of port container based on improved YOLOv5[J]. Foreign Electronic Measurement Technology, 2024, 43(2):16-25. DOI:10.19652/j. cnki.femt. 2305518.

[10] Hou Q, Zhou D, Feng J. Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2021: 13713-13722.

[11] Varghese R, Sambath M. YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness[C]//2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). IEEE, 2024: 1-6.

[12] 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[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2023: 7464-7475. 

[13] Khanam R, Hussain M. What is YOLOv5: A deep look into the internal features of the popular object detector[J]. arxiv preprint arxiv:2407.20892, 2024.

[14] Wang A, Chen H, Liu L, et al. Yolov10: Real-time end-to-end object detection[C]//IEEE Conference on Computer Vision and Pattern Recognition. Seattle WA, USA: IEEE, https://doi.org/ 10.48550/ arXiv.2405.14458.