Automatic Traffic Sign Identification System for Real Time Operation

The traffic sign detection, recognition of signs is a vital part of automatic traffic sign identification system. Its main objective is to guide the driver to be aware of the essential traffic information during driving on the road and to be a major part of automated on-road driving system. The aim of the proposed work is to detect traffic sign which has illumination variation, partial occlusion and to classify using less computational resources and with reduced time complexity in real time environment. Motion model is designed to perform prediction for estimating the traffic sign with the help of kalman filter. Appearance model in the form of adaptive training based detector is used for the final traffic sign detection.