Using street imagery and crowdsourcing internet marketplaces to measure motorcycle helmet use in Bangkok, Thailand
Bibliographic record
Abstract
INTRODUCTION: The majority of Thailand's road traffic deaths occur on motorised two-wheeled or three-wheeled vehicles. Accurately measuring helmet use is important for the evaluation of new legislation and enforcement. Current methods for estimating helmet use involve roadside observation or surveillance of police and hospital records, both of which are time-consuming and costly. Our objective was to develop a novel method of estimating motorcycle helmet use. METHODS: Using Google Maps, 3000 intersections in Bangkok were selected at random. At each intersection, hyperlinks of four images 90° apart were extracted. These 12 000 images were processed in Amazon Mechanical Turk using crowdsourcing to identify images containing motorcycles. The remaining images were sorted manually to determine helmet use. RESULTS: After processing, 462 unique motorcycle drivers were analysed. The overall helmet wearing rate was 66.7 % (95% CI 62.6 % to 71.0 %). Taxi drivers had higher helmet use, 88.4% (95% CI 78.4% to 94.9%), compared with non-taxi drivers, 62.8% (95% CI 57.9% to 67.6%). Helmet use on non-residential roads, 85.2% (95% CI 78.1 % to 90.7%), was higher compared with residential roads, 58.5% (95% CI 52.8% to 64.1%). Using logistic regression, the odds of a taxi driver wearing a helmet compared with a non-taxi driver was significantly increased 1.490 (p<0.01). The odds of helmet use on non-residential roads as compared with residential roads was also increased at 1.389 (p<0.01). CONCLUSION: This novel method of estimating helmet use has produced results similar to traditional methods. Applying this technology can reduce time and monetary costs and could be used anywhere street imagery is used. Future directions include automating this process through machine learning.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 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".