A novel approach for diagnosing road safety issues using automated computer vision techniques
Bibliographic record
Abstract
The use of traffic conflicts for safety diagnosis has been gaining acceptance as a surrogate for collision data analysis. The traffic conflicts approach provided better understanding of collision contributing factors and the failure mechanism that leads to road collisions. This paper demonstrates an automated proactive safety diagnosis approach for vehicles, pedestrian and cyclists using video-based computer vision techniques. Traffic conflicts are automatically detected and several conflict indicators such as Time to collision (TTC) are calculated based on the analysis of the road-user positions in space and time. Additionally, spatial violations are detected based on the non-conformance of road users to travel regulations. Several case studies are described. The first case study deals of the safety analysis of a newly installed bike lane at the southern approach of a heavy volume Bridge in Vancouver, British Columbia. The results showed a high exposure of cyclists to traffic conflicts. Practical solutions to address the safety issues at the location were presented. The second case study deals with a major signalized intersection where concerns were raised regarding the high conflict rate between vehicles and pedestrians as well as an elevated number of traffic violations (i.e., jaywalking) at the intersection. Conflicts and violations were identified and pedestrian safety countermeasures were presented. The third case study deals with a before-after (BA) safety evaluation of a newly proposed design for channelized right-turn lanes at intersections. The new design, termed “Smart Channels”, decreases the angle of the channelized right turn to approximately 70 degrees. The results of the evaluation show that the implementation of the right-turn treatment has resulted in a considerable reduction in the severity and frequency of merging, rear-end, and total conflicts. Overall, the three case studies demonstrate the considerable potential of using video-based computer vision techniques for automated road safety analysis based on traffic conflicts.
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 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".