A Benchmark Dataset for Vision-based Bridge Traffic Load Monitoring in a Cable-stayed Bridge
Bibliographic record
Abstract
Traffic load monitoring based on deep learning and computer vision has garnered significant attention in bridge engineering worldwide. Unlike traditional traffic load monitoring systems, computer vision-based techniques can accurately extract the spatiotemporal load distribution across the entire bridge in an autonomous manner. However, many of the related studies in the literature used datasets that were collected from a few specific areas of different bridges, and there are very limited datasets that provide complete coverage of the entire bridge, making a detailed comparison of different methods difficult. This paper presents a benchmark dataset that provides a series of annotations and field measurements required for traffic load detection, tracking and continuous monitoring on the bridge. The dataset was collected by five cameras, and two weigh-in-motion systems installed on a cable-stayed bridge and is divided into three subsets. The first subset contains over 32,000 images and annotation files of eleven types of vehicle-related targets, which are necessary for the training of vehicle detection models. The second subset consists of photos of the calibration board and coordinates of reference points that are used for camera calibration. The last subset is designated for the field verification of various algorithms, providing synchronized vehicle weight data and monitoring videos covering the whole bridge. To the author’s knowledge, this dataset is the first open-source dataset for vision-based traffic load monitoring in a bridge, which will have tremendous value in promoting research in the area of innovative bridge health monitoring technologies.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.003 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.004 | 0.002 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.006 |
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 source (direct Gemma or distilled Codex), 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".