To tackle the complex issue of open-environment object detection in railway catenaries, characterized by diverse foreign object categories and varying operational scenarios, a foreign object detection method for railway catenaries based on an improved DINO model is proposed. Firstly, by leveraging the image characteristics of foreign object in railway catenaries, the EfficientNet network is employed to replace the ResNet backbone in the original model, further enhancing the Convolutional Block Attention Module (CBAM). Additionally, the neck structure incorporates the enhanced Weighted Bidirectional Feature Pyramid Network (BiFPN) to improve the model’s focus on critical features and enhance detection performance. Secondly, various data augmentation techniques, such as mosaic data augmentation and environmental disturbances, are employed to process the input image data, enriching the features of the sample data. Finally, the application of foreign object detection in railway catenaries is realized through a railway artificial intelligence platform. The results indicate that the proposed method excels in performance, achieving a mean Average Precision of 89.87%, outperforming YOLOv5, DETR and the original DINO by 6.40%, 7.31% and 5.75%, respectively. This method meets the requirements for accurate, rapid, and intelligent identification of foreign objects on railway lines, offering vital technical support for the detection of foreign objects in catenaries.
DUANWangwang, TANGPeng, JINWeidong, et al. Bird's Nest Detection of Railway Catenary Based on HOG Features in Key Areas [J]. China Railway, 2015 (8): 73-77. in Chinese
JINWeidong, YANGPei, TANGPeng. Double Discriminator Generative Adversarial Networks and Their Application in Detecting Nests Built in Catenary and Semi-Supervized Learning [J]. Scientia Sinica: Informationis, 2018, 48 (7): 888-902. in Chinese
LIUGuowen, ZHANGCaixia, LIBin, et al. Detection of Bird's Nest in Catenary Based on Improved RetinaNet Model [J]. Journal of Data Acquisition and Processing, 2020, 35 (3): 563-571. in Chinese
[7]
吕嘉宜.基于视觉的高速铁路接触网异物检测[D].杭州:浙江大学,2021.
[8]
JiayiLÜ. Vision-Based Detection of Foreign Bodies in Catenary of High-Speed Railway [D].Hangzhou: Zhejiang University, 2021. in Chinese
[9]
顾(龙天)楠.铁路接触网异物检测图像处理技术研究[D].兰州:兰州交通大学,2022.
[10]
GULongtian) Nan. Research on Image Processing Technology of Foreign Body Detection in Railway Catenary [D]. Lanzhou: Lanzhou Jiaotong University, 2022. in Chinese
WANGXiaohong, DUYunfei, LIUChang. Contact Line Bird Nest Foreign Object Detection Based on YOLOV5s [J]. Yangtze River Information Communication, 2023, 36 (6): 51-54. in Chinese
WANGHongliang, LIUJiajun, YIHang, et al. The Foreign Objects Removal System of the Traction Power Supply Catenary Based on Machine Vision [J]. Power System and Clean Energy, 2021, 37 (11): 72-79. in Chinese
ZHOUMing, WANGJijun, RONGZhengguan, et al. Research on Early Warning and Monitoring Scheme for Suspended Floating Objects in the High-Speed Railway Overhead Contact System [J]. Railway Construction Technology, 2023 (11): 25-27, 139. in Chinese
[17]
ZHANGH, LIF, LIUS L, et al. DINO: DETR with Improved De-Noising Anchor Boxes for End-to-End Object Detection [J]. ArXiv e-Prints, 2022: arXiv:
CAOChunsheng, WANGWenhao, XIEDapeng, et al. OCS Safety Patrol Inspection System Based on Image Recognition [J]. Railway Technical Innovation, 2023 (6): 50-54. in Chinese
XUWei, WUZebin, LIUJianxin, et al. Intelligent Foreign Objects Detection Method for OCS of High Speed Railway [J]. China Railway, 2019 (10): 39-44. in Chinese
[22]
ZHUX Z, SUW J, LUL W, et al. Deformable DETR: Deformable Transformers for End-to-End Object Detection [EB/OL]. 2020: arXiv: 2010.04159.
JIANGXinlan, JIAWenbo. Machine Vision Detection Method for Foreign Object Intrusion in High-Speed Rail Contact Net [J]. Computer Engineering and Applications, 2019, 55 (22): 250-257. in Chinese
[27]
LIUS L, ZENGZ Y, RENT H, et al. Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection [J]. ArXiv e-Prints, 2023: arXiv:
WANGKeli, GAOFulai, YANGPeng, et al. Recognition and Detection of Catenary Bird's Nest Based on Deep Learning [J]. Railway Locomotive & Car, 2022, 42 (2): 116-121. in Chinese
SHITianyun, HOUBo, LIGuohua, et al. Design and Key Technology Research of Railway Artificial Intelligence Platform [J]. Railway Computer Application, 2023, 32 (8): 9-16. in Chinese
[32]
ZHAOY A, LÜW Y, XUS L, et al. DETRs Beat YOLOs on Real-Time Object Detection [EB/OL]. 2023: arXiv: 2304.08069.
[33]
LINT Y, DOLLÁRP, GIRSHICKR, et al. Feature Pyramid Networks for Object Detection [C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, HI, USA. New York: IEEE Press, 2017: 936-944.
[34]
LIUY C, SHAOZ R, HOFFMANNN. Global Attention Mechanism: Retain Information to Enhance Channel-Spatial Interactions [EB/OL]. 2021: arXiv: 2112.05561.
[35]
WOO S, PARKJ, LEEJ Y, et al. CBAM: Convolutional Block Attention Module [C]// European Conference on Computer Vision. Munich, Germany. Cham: Springer, 2018: 3-19.