Looking-In and Looking-Out of a Vehicle: Computer-Vision-Based Enhanced Vehicle Safety

In this article the authors present research into real-time machine-vision and driver-support systems for intelligent vehicles. For such systems, researchers put forward a tripartite scheme of concern: 1) the environment, which is the roadway surface and infrastructure as well as climactic situations; 2) the vehicle, which, in this situation, is to be equipped with telematics devices for driving assistance; and, 3) the driver. This scheme, Environment, Vehicle, and Driver (EVD), is used in a framework for the multifunctional active safety context-capture system. The computer vision component of the system is examined not just for external object detection, but also for cues from the driver in order to better predict vehicular movement. The authors propose that, for a truly active multimodal sensory approach, a diverse array of sensors to be fused would be required.

  • Availability:
  • Authors:
    • Trivedi, Mohan Manubha
    • Gandhi, Tarak
    • McCall, Joel
  • Publication Date: 2007-3


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01044882
  • Record Type: Publication
  • Source Agency: UC Berkeley Transportation Library
  • Files: BTRIS, TRIS
  • Created Date: Mar 22 2007 2:31PM