Plate Detection and Recognition by using Colour Information and ANN

Abstract

Automatic car plate recognition finds large application areas in the contemporary world It can be used in carparks and/or places where entries and exits are wanted under control. A car plate recognition system consist of three main division. These are; detecting the plate location, segmenting the characters and recognizing the characters. Processing the whole picture in order to determine plate location takes lots of time. Instead of processing whole picture, determining candidate plate locations and processing these areas increase achievement of the application and takes less time consuming. Plates background and character's colour are used to determine candidate plate locations. Character segmentation is achieved with calculating column sum vector and artificial neural network (ANN) is used to recognise characters.