Occupant classification system for automotive airbag suppression
CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition
Hi-index | 0.00 |
When an airbag deploys on a rear-facing infant seat, itcan injure or kill the infant. When an airbag deploys on anempty seat, the airbag and the money to replace it arewasted. We have shown that video images can be used todetermine whether or not to deploy the passenger-sideairbag in a crash. Images of the passenger seat, taken froma video camera mounted inside the vehicle, can be used toclassify the seat as either empty, containing a rear-facinginfant seat, or occupied. Our first experiment used asingle, monochrome video camera. The system wasautomatically trained on a series of test images. Using aprinciple components (eigenimages) nearest neighborclassifier, it achieved a correct classification rate of 99.5%on a test of 910 images. Our second experiment used apair of monochrome video cameras to compute stereodisparity (a function of 3D range) instead of intensityimages. Using a similar algorithm, the second approachachieved a correct classification rate of 95.1% on a test of890 images. The stereo technique has the advantage ofbeing less sensitive to illumination, and would likely workbest in a real system.