Evaluation of Camera-Based Systems to Reduce Transit Bus Side Collisions: Phase II

Lin, Pei-Sung; Kourtellis, Achilleas; Wills, Matthew · 2012 · ROSA P / Florida. Dept. of Transportation. Research Center

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 evaluates an integrated camera-mirror hybrid system designed to eliminate blind zones and reduce side collisions in transit buses. Side collisions account for over 40% of transit crashes, and traditional mirrors fail to cover critical areas around the vehicle. While Phase I of this project demonstrated the potential of aftermarket wide-angle cameras, those systems provided excessive visual information that distracted drivers. Consequently, Phase II focused on developing a refined system with a narrower 65-degree horizontal field of view, which adequately covers blind zones without overwhelming the operator. The research aimed to design this integrated system, deploy it in field settings, analyze performance data, and develop specifications for widespread industry adoption. The researchers developed two distinct systems for different bus types. For Type A buses, they utilized Dallmeier cameras housed in custom, nitrogen-filled PVC enclosures to prevent fogging, paired with 7-inch and 8-inch LCD monitors. For cutaway buses, they employed a commercial VELVAC mirror-camera integration system. Evaluation involved a controlled driving test with 29 transit drivers at StarMetro in Tallahassee, Florida. Drivers performed object detection tasks using traffic cones and human subjects placed at random locations around static buses. Performance was measured by comparing accuracy and response times between using mirrors alone versus the hybrid camera-mirror system. Following the controlled test, the systems underwent a longer field deployment with LeeTran in Fort Myers to assess real-world usability and durability. Results from the controlled driving test indicated significant improvements in object identification. Drivers using the camera system achieved a 96–98% correct identification rate for object location, compared to 70–78% when using mirrors only. Surprisingly, drivers identified objects faster with the camera system despite having two additional search locations (the monitors) to check. Driver feedback was overwhelmingly positive, with most agreeing that the system effectively eliminated blind zones. During the field deployment, drivers reported that the system helped reduce side collision risks. However, issues such as initial fogging (resolved via nitrogen filling) and distraction from passing vehicle headlights at night were noted. The nighttime glare issue was not fully resolved, though the cameras selected had features to mitigate blooming. The study concludes that integrated camera-mirror systems are effective tools for enhancing transit bus safety by eliminating blind zones. The findings support the development of specific technical specifications for monitors, cameras, and housing to guide practitioners and fleet managers. While the system proved superior to mirrors in accuracy and speed, the research highlights the need for further investigation into nighttime glare mitigation to ensure the system does not introduce new hazards. The compiled recommendations provide a framework for selecting and implementing these systems in transit fleets.

Key finding

Drivers achieved 96-98 percent correct object identification and responded faster using the camera-mirror hybrid system compared to 70-78 percent accuracy with mirrors only.

Methodology

mixed_methods

Sample size: 29

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).

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.

• automated enforcement cameras