Promoting excellence in mobility engineering

  1. FISITA Store
  2. Technical Papers

Real-time Visual Recognition and Scene Interpretation for the Intelligent Vehicle
ADAS2002/02A4012

Authors

Jörg-Michael Schneider - Philips Semiconductors

Abstract

The Presentation shows a new System-on-Chip (SoC) design for specific applications such as "adaptive cruise control” and other driver assistance applications like lane departure warning systems, forward collision warning, and sensory fusion for increased roadway safety. The system is using a single CMOS image sensor and the SoC chip to interpret and detect other vehicles, pedestrians, road signs and other objects while providing driver assistance.

The SoC architecture contains multiple ARM946 RISC cores and four application-specific modules for image pre-processing, motion analysis, pattern recognition, and lane following on the road.

The SoC will include proprietary pattern identification to discriminate between vehicles and background scenery. It will also include visual motion analysis methods for isolating dynamically-moving patterns, such as passing and crossing vehicles; software for estimating a vehicle's yaw and pitch rates; and image processing techniques for lane-following and path prediction.

The chip is geared toward camera-based adaptive cruise control .

Add to basket

Back to search results