Pulse coded neural network for sign recognition for navigation

Abstract

In this paper we present a sign recognition for automated machines (such as robots and vehicles). This system is also suitable for assisting navigators. This new method uses a database of binary barcodes for sign recognition. Binary barcodes are generated from images of signs using a pulse coded neural network (PCNN). Matching the barcode of an image with the barcodes in the database leads to automatical identification of the unknown detected sign. The method is fast, reliable and easy to implement compared to the known sign recognition methods.