Deep Learning-Based Railway Foreign Object Intrusion Intelligent Perception Using Attention-Aggregated Semantic Segmentation
- 16 October 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 30 (4) , 2609-2619
Abstract
Foreign object intrusion detection (FOID) is one of the critical tasks to ensure the safe and efficient operation of trains. Semantic segmentation, which involves pixel-level recognition of images, has been widely studied in automatic driving obstacle avoidance. However, unlike road transportation, the operation speed of trains requires higher detection efficiency. The availability of mature railway scenario datasets is limited compared to road transportation datasets. Therefore, considering the complexity of operating scenarios with diverse and unpredictable foreign objects, this article proposes a boundary-assisted dual-branch attention semantic segmentation network (BDANet). BDANet completes accurate segmentation while reducing parameters, enabling real-time semantic recognition of the railway environment. A COCO-Stuff-Rail dataset extracted based on COCO-Stuff is constructed to guide model training. Then, an adaptive correction algorithm is introduced to fine-tune the BDANet, making it generalizable to diverse realistic environments. Ultimately, this article achieves end-to-end track extraction, open-set foreign object detection, and common foreign object identification using a unified process. To evaluate the superiority of BDANet, comparison, and ablation experiments are conducted on the COCO-Stuff-Rail. Visual segmentation and open-set detection results of a real-world scenario validate that the proposed process can bridge the gap between the training set and practical applications.Keywords
Funding Information
- National Natural Science Foundation of China (62327806)
- National Natural Science Foundation of China (61925302)
- National Natural Science Foundation of China (62273027)
- Natural Science Foundation of Beijing Municipality (L211021)
- Open Project of Shanghai Driverless Train Control System Engineering Technology Research Center (GrantSUTC-2024KT-05)
This publication has 39 references indexed in Scilit:
- Quadratically Constrained Linear Programming-based energy-efficient driving for High-speed Trains with neutral zone and time windowTransportation Research Part C: Emerging Technologies, 2023
- Application and Evaluation of Image-based Information Acquisition in Railway TransportationJournal of Intelligent & Robotic Systems, 2022
- Urban rail transit obstacle detection based on Improved R-CNNMeasurement, 2022
- An effective railway intrusion detection method using dynamic intrusion region and lightweight neural networkMeasurement, 2022
- Multi-scale object detection for high-speed railway clearance intrusionApplied Intelligence, 2021
- Energy‐efficient train operation with steep track and speed limits: A novel Pontryagin's maximum principle‐based approach for adjoint variable discontinuity casesIET Intelligent Transport Systems, 2021
- Artificial Intelligence for Obstacle Detection in Railways: Project SMART and Beyond2020
- Multi-block SSD based on small object detection for UAV railway scene surveillanceChinese Journal of Aeronautics, 2020
- Unified Perceptual Parsing for Scene Understanding2018
- Railway obstacle detection algorithm using neural networkAIP Conference Proceedings, 2018