Stereovision-based real-time occupant classification system for advanced airbag systems

Abstract

Occupant classification in a passenger seat is one of the critical components for any advanced airbag system. Many automotive electronic suppliers and engineers predict that the camera will be the next generation sensor for active and passive safety systems because it has several advantages compared to other sensors. The present paper describes a stereovision-based occupant classification system (OCS) and intelligent algorithm with embedded system by which triggering of the airbag deployment can be controlled. The system consists of a pair of stereo cameras and dual Digital Signal Processor (DSP): the first DSP is for the stereo matching processing, and the second is for occupant classification. The results show that the reaches 97%, and the processing time is 960 ms. Such performance indicates that the feasibility of the system as an embedded OCS is high.