Model-based target recognition in pulsed ladar imagery

A pulsed ladar based object-recognition system with applications to automat ic target recognition (ATR) is presented. The approach used is to fit the s ensed range images to range templates extracted through a laser physics bas ed simulation applied to geometric target models. A projection-based prescr eener filters out more than 80% of candidate templates. For recognition, an M of N pixel matching scheme for internal shape matching is combined with a silhouette matching scheme, The system was trained on synthetic data obta ined from the simulation, and has been blind tested on a data set containin g real ladar images of military vehicles at various orientations and ranges . Succesful blind testing on real imagery demonstrates the utility of synth etic imagery for training of recognizers operating on ladar imagery.