Details:

Creators:
Lin, Pei-Sung ; Chen, Cong ; Ozkul, Seckin ; Rajalingola, Manvitha
Lin, Pei-Sung ; Chen, Cong ; Ozkul, Seckin ; Rajalingola, Manvitha Less -
Corporate Creators:
University of South Florida. Center for Urban Transportation Research
Corporate Contributors:
Florida. Department of Transportation
Subject/TRT Terms:
[+]
Detection And Identification Systems Human Factors In Crashes Testing Traffic Control Centers Traffic Safety Wrong Way Driving
Publication/ Report Number:
BDV25-977-40
Resource Type:
Tech Report
Geographical Coverage:
United States ; Tampa (Florida)
United States ; Tampa (Florida) Less -
Edition:
Final Report, 4/1/17–11/30/18
Corporate Publisher:
Florida. Department of Transportation ; University of South Florida. Center for Urban Transportation Research
Florida. Department of Transportation ; University of South Florida. Center for Urban Transportation Research Less -
Abstract:
Although wrong-way driving (WWD) crashes account for a relatively small portion of total crashes, the impact between two cars crashing into each other at high speeds in opposite directions causes a tremendous amount of damage compared to any other types of crashes, especially on freeways or other limited access facilities. Therefore, there is a continuing need to expand opportunities to reduce crashes by applying new technologies. Sponsored by Florida Department of Transportation (FDOT), this project successfully evaluated video-analytic freeway WWD detection systems currently on the market from three vendors, regarding their capabilities for real-time WWD vehicle detection and Traffic Management Center (TMC) notification. In this project, in addition to actual WWD vehicles, traffic of inside-lane driving in the opposite direction (IDOD) was treated as WWD traffic under certain data collection scenarios for WWD detection to ensure sufficient data size. Six testing locations were selected on I-275 segment between I-4 and I-75 in the Tampa Bay area, and these testing locations were assigned to one of four testing scenarios: (1) testing with normal daily traffic conditions; (2) testing consecutive WWD in both directions; (3) testing under normal light nighttime traffic conditions; and (4) testing under low light nighttime traffic conditions. Real-time WWD incident detection and data collection were conducted through fixed camera videos streams at these locations. A series of performance measures was defined to evaluate the performance of the WWD detection systems from three vendors, including (1) WWD detection system accuracy; (2) percentage of false calls; (3) actual WWD detection accuracy; and (4) percentage of missed calls. In addition, the capabilities of candidate systems on TMC notification were tested by collecting email notification data. The data review and analysis results revealed that (1) WWD is a very rare traffic incident, as there were no actual WWD vehicles detected in real traffic during the one-week data collection period; (2) in all three testing scenarios, including testing of consecutive WWD vehicles in both directions, testing under normal light nighttime traffic conditions, and testing under low light nighttime traffic conditions, researchers found that the system from Vendor 1 showed the best performance at an overall 95% detection system accuracy and 94% actual detection accuracy, followed by the systems from Vendor 2 and Vendor 3; (3) system performance from the three vendors on WWD detection varied significantly, indicating that performance of real-time video-analytic freeway WWD detection systems highly depends on the individual vendor’s system; (4) analysis of TMC notification data revealed that all three candidate systems were able to send an email notification if a WWD was detected. The evaluation results and findings from this research project are informative and can be used to support FDOT and other state DOTs in future implementations of WWD detection systems on freeways and limited-access facilities.
Format:
PDF
Funding:
BDV25-977-40
Collection(s):
US Transportation Collection
Main Document Checksum:
[+]
urn:sha256:714f2231aea6632dada8c027927e2a372abe9641432afcbb1cb40fc3f920d261
Download URL:
https://rosap.ntl.bts.gov/view/dot/60056/dot_60056_DS1.pdf
File Type:

Supporting Files

• No Additional Files

More +