Application of a Stereovision Sensor for the Occupant Detection and Classification in a Car Cockpit
Antonio Marı́n-Hernández, Michel Devy · 2000
In this paper, we present recent methods that we have proposed to deal with the occupant detection and classification in a vehicle. The aim of this work is to provide information about the passenger presence and location within the vehicle, from stereo data. The results of this classification will be used for an active airbag control: the inflation could be activated, depovered or deactivated with respect to the seat configuration. The paper will focus on the extraction of a 3D global representation of the seat area, from a 3D image, and on the classification performed using this global description.