Railway perimeter intrusion pedestrian detection algorithm based on trajectory collaboration
ZHU Baoquan
CHEN Weijian
LI Xiaozheng
ZHANG Ning
ZOU Sijie
WANG Yuteng
GUO Baoqing
Abstract：Pedestrian intrusion into railway perimeters poses a serious threat to train safety.To address the high false-negative rate of single-image pedestrian detection algorithms based on deep learning in complex railway environments,this paper proposes a railway perimeter intrusion detection algorithm that combines YOLOv5x single-frame detection results with pedestrian trajectory correlation in image sequences.First,high-confidence pedestrian targets from single-frame images are integrated with three types of features,including spatial relationships of trajectories,trajectory irregularity informa-tion,and matching results between detections and trajectory predictions,to achieve reliable pedestrian trajectory updating and filtering.Second,the trajectory-correlated prediction results are used to reset the confidence scores of low-confidence pedestrian targets in single-frame images,thereby improving detection performance in challenging railway environments such as nighttime and adverse lighting con-ditions.Furthermore,a pedestrian intrusion dataset covering five typical railway scenarios is con-structed,including cases prone to missed detections,such as occluded objects,blurred targets,and distant small objects under low-light or glare conditions.Finally,to further validate the effectiveness of the proposed algorithm,Faster R-CNN and YOLOv3 are employed as detectors to assess the per-formance improvement.Experimental results demonstrate that,compared to the single-frame detec-tion algorithm YOLOv5x,the proposed method improves recall by 4.5%and reduces the log-average miss rate by 4.0%without significant loss in detection precision,proving the efficacy of the trajectory updating,filtering,and confidence reset modules.Additionally,compared to Faster R-CNN and YO-LOv3,the proposed algorithm achieves recall improvements of 2.3%and 3.6%,respectively,and re-duces log-average miss rates by 1.3%and 3.5%,confirming its adaptability to different detection models.
Keywords：intelligent transportationrailway pedestrian intrusiontrajectory state characteristictra-jectory collaborationconfidence reset
Publication Date：2025-12-30
Online Publishing Date：2026-02-02(First online date of this platform, not the publication date of the document)
Pages：10( 137-146 )