Testing and Evaluation of Freeway Wrong-way Driving Detection Systems
archive: archived pipeline: cataloged verified
Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)
Summary
This study addresses the critical safety issue of wrong-way driving (WWD) on freeways, where high-speed, head-on collisions cause disproportionate damage despite their low frequency. Motivated by the need for effective real-time detection to alert Traffic Management Centers (TMCs) and law enforcement, the research evaluates the performance of three commercial video-analytic WWD detection systems. The project was sponsored by the Florida Department of Transportation (FDOT) and conducted by the Center for Urban Transportation Research (CUTR) at the University of South Florida. The experimental design involved six testing locations on Interstate 275 in the Tampa Bay area, utilizing fixed cameras to ensure stable regions of interest, as pan-tilt-zoom cameras proved unreliable for maintaining consistent views. Because actual WWD incidents are extremely rare, the researchers employed an innovative data collection strategy: one location monitored actual traffic, while five others used simulated WWD data. Specifically, right-way traffic was treated as WWD at one site, and inside-lane driving in the opposite direction (IDOD) was treated as WWD at the remaining four sites. Data were collected over one week across four scenarios: normal daily traffic, consecutive WWD in both directions, normal light nighttime conditions, and low light nighttime conditions. Performance was measured using detection system accuracy, false call percentage, actual detection accuracy, and missed call percentage, with video reviews serving as ground truth. The results confirmed that actual WWD is exceedingly rare, with no true incidents detected during the week-long normal traffic monitoring. In simulated scenarios, Vendor 1 consistently outperformed the other two vendors. Vendor 1 achieved an overall detection system accuracy of 95% and an actual detection accuracy of 93–94%, depending on the scenario. Vendor 2 ranked second with 73% system accuracy and 50% actual accuracy, while Vendor 3 performed the worst, with only 28% system accuracy and 12% actual accuracy. Statistical analysis indicated that the performance differences between vendors were significant at the 5% level. All three systems successfully transmitted email notifications to the TMC upon detection. The study also noted that while normal lighting provided better illumination, low-light conditions offered better contrast, though no definitive conclusion could be drawn regarding which condition yielded superior detection accuracy without further research. The significance of this research lies in its demonstration that video-analytic WWD detection capabilities vary drastically by vendor, highlighting the importance of rigorous evaluation before deployment. The findings provide FDOT and other state departments of transportation with evidence-based insights to support the implementation of WWD detection systems on limited-access facilities. The study recommends that future implementations consider the specific performance metrics of individual vendors and suggests improvements for PTZ camera integration, such as automatic adjustment for minor movements, to enhance system reliability.
Key finding
Vendor 1's wrong-way driving detection system achieved an overall 95% detection system accuracy and 93% actual detection accuracy, significantly outperforming Vendor 2 and Vendor 3.
Methodology
field_study
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. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | rosap | — | — | 2 | 2026-05-23 |
| archive | success | — | — | — | 1 | 2026-05-23 |
| extract | success | cached | — | — | 2 | 2026-06-10 |
| clean | success | — | — | — | 1 | 2026-06-01 |
| chunk | success | — | — | — | 1 | 2026-06-01 |
| embed | success | — | — | — | 1 | 2026-06-02 |
| enrich | success | — | — | — | 1 | 2026-05-23 |
| promote | success | — | — | — | 1 | 2026-05-23 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 3 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 19 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; 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: validation psychometrics