Track Intrusion Detection and Track Integrity Evaluation – Year 2

Details

• Creators:
  Qian, Yu ; Tang, Youzhi ; Rizos, Dimitris C ; Vitzilaios, Nikolaos
• Corporate Creators:
  University Transportation Center for Railway Safety (UTCRS) [UTC] ; University of South Carolina
• Corporate Contributors:
  United States. Department of Transportation. University Transportation Centers (UTC) Program ; United States. Department of Transportation. Office of the Assistant Secretary for Research and Technology
• Subject/TRT Terms:
  Artificial Intelligence At Grade Intersections Computer Vision Grade Crossing Protection Systems Neural Networks Railroad Safety Trespassers
• Publication/ Report Number:
  UTCRS-USC-O7CY24
• Resource Type:
  Tech Report
• Geographical Coverage:
  United States
• Edition:
  June 1, 2024 – August 31, 2025
• Corporate Publisher:
  University Transportation Center for Railway Safety (UTCRS) Tier-1 University Transportation Center (UTC)
• Abstract:
  Track intrusion, especially trespassing within railroad rights-of-way, poses a major safety risk, causing more fatalities than train-vehicle collisions. Existing methods often struggle with real-time detection on edge devices. This study introduces YOLO-RCNN, a novel model integrating YOLO-FG for comprehensive object detection and segmentation, optimized for real-time, resource-constrained environments. YOLO-RCNN achieves superior performance, with deployment optimizations boosting inference speed from 4.69 FPS to 54.79 FPS on desktop and from 3.39 FPS to 29.19 FPS on Jetson AGX Orin, highlighting its potential for efficient, reliable railroad crossing monitoring.
• Format:
  PDF
• Funding:
  69A3552348340
• Collection(s):
  University Transportation Centers
• Main Document Checksum:
  urn:sha-512:a860886b30b3aaf6b0043c643af425e44c2fb4414a89fd4cd1cbec724cd21c4eaf0871645b3d0946b647414ea18af0b599ac95c66651727f4d614ee91256eb3a
• Download URL:
  https://rosap.ntl.bts.gov/view/dot/88092/dot_88092_DS1.pdf
• File Type:
  [PDF - 1.83 MB ]