Object tracking and classification for multiple active safety and comfort applications using a multilayer laserscanner
Kay Ch. Fuerstenberg, Klaus Dietmayer · 2004
This paper summarises new approaches combining distance and reflectivity measurements of laser scanners in order to track and classify more accurately. Additionally a new method is introduced, which analyses the motion of the legs of a pedestrian, in order to distinguish between moving pedestrians and balls or boxes in motion. A first estimation of the motion of the pedestrians legs is performed just one second after the first detection. The analysis is widely independent of the pedestrians' distance and direction of motion.