Research on Machine Vision–Based Intelligent Tracking System for Maintenance Personnel
Bibliographic record
Abstract
Upon returning to the depot, rail transit vehicles require necessary maintenance. The working condition of train maintenance personnel directly impacts the safety of both staff and equipment. Therefore, effective monitoring and control of activities within train roof access platforms are essential. Traditional manual monitoring demands substantial manpower and is prone to human error, whereas machine vision–based intelligent monitoring offers a promising alternative, reducing the dispatch control center (DCC) workload while enhancing safety management. Our intelligent monitoring approach involves three key steps: train maintenance personnel identification, tracking of maintenance activities to generate movement trajectories, and analysis of movement patterns to detect anomalous behavior. This study primarily addresses the challenges of personnel identification and process tracking. In the scenario of train maintenance, facial recognition is limited by posture variations, making direct video tracking impractical. Pedestrian reidentification (Re‐ID) also struggles with posture and attire changes. To address these issues, we propose a hybrid approach: facial recognition confirms personnel identity upon entry, followed by pedestrian feature extraction for Re‐ID‐based tracking throughout the maintenance process. To handle occlusion, we designed a Re‐ID method based on body part recognition, segmenting features into head–shoulder, body, arm, and leg components, with higher weights assigned to visible parts. This method achieved improved mean average precision (mAP) and Rank‐1 values of 87.6% and 95.7%, respectively, on the Market1501 dataset. A tracking and monitoring system was developed, effectively identifying and tracking maintenance activities, demonstrating a strong practical value. Furthermore, this work lays the groundwork for future research into trajectory‐based abnormal behavior detection.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".