Multiclass Vehicle Type Recognition System using an Oriented Points Set based Model

PABLO NEGRI; XAVIER CLADY; LIONEL PREVOST; MAURICE MILGRAM; RAPHAEL POULENARD

Tunez

Conferencia; IEEE International Multi-Conference on Systems, Signals & Devices; 2007

This paper presents a framework for multiclass vehicle type identification based on oriented contour points. The decision process combines four classifiers: three voting algorithms and a distance error. This method have been tested on a realistic data set obtaining similar results for equivalent recognition frameworks with different features selections. A confidence criterion is also calculated to validate the system results. The system also show to be robust to partial occlusions.