A Comprehensive Labeling Protocol and Real-time Inference Framework for Road Hazard Detection
Bibliographic record
Abstract
Real-time detection of road surface damage and hazards represents a critical component in intelligent transportation systems and autonomous driving applications. However, existing methodologies often fail to capture the complexity and variability of real-world driving environments, while public datasets suffer from inconsistent labeling protocols and limited robustness to visual ambiguities. This paper introduces a comprehensive framework that addresses these challenges through: (1) a systematic annotation protocol optimized for real-world driving conditions, and (2) an edge-server hybrid inference architecture for efficient deployment. We collected 13,000 high-resolution road scene images and systematically re-annotated multiple public datasets to create a high-quality training corpus of 595,530 images. Visual artifacts such as shadows, surface contamination, and worn paint markings were systematically excluded through rigorous annotation guidelines. The proposed lightweight YOLO-based detector, deployed within a hybrid edge-server pipeline, achieves low-latency, high-throughput inference while maintaining high detection accuracy. Comprehensive evaluations on large-scale real-world benchmarks demonstrate that our approach reduces server load by 77%, achieves 23.0 FPS inference speed, and attains a competitive mAP of 0.7147, confirming its suitability for deployment in safety-critical in-vehicle systems.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.004 |
| Open science | 0.004 | 0.003 |
| Research integrity | 0.001 | 0.003 |
| Insufficient payload (model declined to judge) | 0.004 | 0.004 |
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".