A System for Traffic Violation Detection
DOI: 10.3390/s141122113
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This paper presents an experimental framework for an Advanced Driver Assistance System (ADAS) designed to detect traffic violations and provide drivers with feedback to improve safety. Motivated by the high incidence of accidents caused by driver inattention and unintentional violations such as speeding, the system aims to alert drivers to road conditions and record infractions to encourage behavioral change. The platform integrates a computer vision subsystem for traffic sign detection and recognition with an Event Data Recorder (EDR) that logs violation data, including GPS coordinates and vehicle speed, for later visualization via Google Earth. The experimental test-bed is installed in a Nissan Note and comprises a Mini-ITX host computer and a slave PC for real-time image processing. The vision subsystem utilizes two roof-mounted cameras: one for daytime and one for nighttime, the latter employing active near-infrared (NIR) illumination to enhance sign reflectivity and reduce motion blur. Vehicle data, including speed and steering angle, is acquired via the OBD-CAN interface and external sensors. Driver identification is managed through smart cards, allowing personalized violation records. The Traffic Sign Detection and Recognition (TSDR) module processes images through dynamic shutter adjustment, color segmentation, shape analysis, and pattern matching. For nighttime, image subtraction techniques isolate reflective signs. The system currently detects three specific violations: speeding, ignoring stop signs, and forbidden turns. Testing involved 2,000 kilometers of driving across various conditions. The TSDR module operates at 4 frames per second, providing sufficient detection opportunities at normal speeds. In clear daytime conditions, the system achieved a 90% true positive detection rate for relevant signs, with performance improving to 91.5% under nighttime conditions due to easier segmentation. False positive rates remained low, under 2%. The system successfully distinguished between similar speed limit signs (e.g., 40, 60, 80, 90) through adaptive thresholding. Alerts are issued with sufficient lead time—approximately 1–2 seconds for speeding at 100 km/h and 3–4 seconds for urban violations—to allow driver reaction. If the driver fails to comply, the violation is recorded with associated imagery and location data. The study concludes that the proposed system effectively detects traffic signs and records violations in real-time, offering a viable method for providing drivers with spatial and temporal feedback on their driving behavior. While currently an experimental platform, the framework demonstrates potential for reducing unintentional infractions and enhancing road safety. Future work includes expanding detection capabilities to additional violation types and conducting qualitative studies with traffic agencies to assess driver perception and system efficacy.
Provenance
The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | Crossref | — | — | 1 | 2026-06-19 |
| archive | success | openalex | — | — | 5 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-20 |
| chunk | success | chunk | — | — | 1 | 2026-06-20 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-20 |
| promote | success | — | — | — | 1 | 2026-06-19 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-20 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Methodological Resource: tool software