A station dense pedestrian tracking system and method based on spatial weak clues
The station dense pedestrian tracking system based on spatial weak cues utilizes a lightweight spatiotemporal attention mechanism and a modified Kalman filter module, combined with weak cues, to achieve efficient pedestrian tracking, solving the detection and tracking challenges in dense pedestrian environments and realizing high-precision and fast pedestrian analysis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHEASTERN UNIV CHINA
- Filing Date
- 2024-02-01
- Publication Date
- 2026-06-19
AI Technical Summary
Traditional methods struggle to efficiently track pedestrians in densely populated environments like subway stations, especially when occlusion and clustering occur, leading to detection and tracking failures. Existing technologies lack effective utilization of cross-frame information and spatial weak cue enhancement.
A station-based dense pedestrian tracking system based on weak spatial cues is adopted. It extracts target features through a lightweight spatiotemporal attention mechanism, and combines modified Kalman filtering and trajectory management modules. It uses weak cues such as trajectory confidence, mixed intersection-over-union ratio, and velocity direction to perform high-confidence and low-confidence correlation to achieve accurate pedestrian tracking.
It improves the accuracy and speed of pedestrian tracking, enabling real-time detection and analysis of dense crowds in complex environments, and enhances the monitoring capabilities for subway station safety.
Smart Images

Figure CN118037774B_ABST
Abstract
Citation Information
Patent Citations
Pedestrian multi-target tracking method and device and computer readable storage medium
CN115240130A
Multi-target tracking method in dense scene
CN116883452A