Automating Wrong Way Detection Using Existing CCTV Cameras

Wrong way driving has been defined as “vehicular movement along a travel lane in a direction opposing the legal flow of traffic”. Although wrong way driving crashes occur infrequently accounting for almost 3 percent of all crashes, they have a very high likelihood of resulting in fatal or serious injury crashes. Vaswani found that wrong-way crashes had 27 times higher fatality rates as compared to any other kind of crashes occurring on controlled-access highways in Virginia. Another study, from Michigan, found that 22% of all wrong-way crashes resulted in a fatality, as compared to, only 0.3% percent of all highway crashes resulted in fatality for the same time frame. The main causes of wrong way driving constitute of: (a) Alcohol- In an analysis conducted on FARS database, from 2002-2009, it was found that 60% of wrong way crash drivers had some indications of alcohol involvement; (b) Age- Driver over age of 70 constitute nearly 15 percent of wrong way drivers as compared to only 3 percent of the right way drivers involved in wrong way crashes; (c) Poorly marked ramps- The primary origin was for controlled-access highway was found to be the exit-ramps; (d) Time of day- Disproportionate amount of the wrong-way fatalities happen during night time. The causes associated with wrong way crashes tend to make them spatially concentrated to particular stretches of roads, thus making it important to identify and monitor such high-risk locations. FHWA Highway Safety Improvement Program recommends generating wrong way monitoring warrants based on total collision and fatal collision rates. If either total wrong-way collision are greater than 0.5 event per mile per year, or a fatal wrong-way collision rate of 0.12 per mile per year and at least 3 wrong-way collisions are recorded for a five-year period, the monitoring warrants are met. This approach is reactive and needs the crash history to develop for a period of at least 3 to 5 years. In this research, the research team proposes to use a pro-active technology that count the number of wrong-way drivers that are detected using image based technology. The proposed solution is discussed in detail in subsequent sections.

    Language

    • English

    Project

    • Status: Active
    • Funding: $60000
    • Contract Numbers:

      Add 677

      TSF-000--(448)--92-00

    • Sponsor Organizations:

      Iowa Department of Transportation

      800 Lincoln Way
      Ames, IA  United States  50010
    • Managing Organizations:

      Iowa State University, Ames

      Center for Transportation Research and Education
      2711 South Loop Drive, Suite 4700
      Ames, IA  United States  50010-8664
    • Project Managers:

      Worrel, Brian

    • Performing Organizations:

      Iowa State University, Ames

      Center for Transportation Research and Education
      2711 South Loop Drive, Suite 4700
      Ames, IA  United States  50010-8664
    • Principal Investigators:

      Sharma, Anuj

    • Start Date: 20181001
    • Expected Completion Date: 20200731
    • Actual Completion Date: 0

    Subject/Index Terms

    Filing Info

    • Accession Number: 01702879
    • Record Type: Research project
    • Source Agency: Iowa Department of Transportation
    • Contract Numbers: Add 677, TSF-000--(448)--92-00
    • Files: RiP, STATEDOT
    • Created Date: Apr 24 2019 2:43PM