Recognition Number Plate Using ACA for Improved Segmentation and Classification

In this paper a number plate recognition system which has been designed using the ant colony optimization technique. This system can be implemented in surveillance systems, detection of stolen vehicles and checking of vehicles at toll plazas, posts, barriers sand other entry points. This research is focusing, an ant colony based number plate extraction method is proposed. Ant colony optimization technique gives better results in edge detection while applying image segmentation. So Better accuracy can be achieved by using this concept in number plate recognition. The natural behavior of ant species that the ant deposit pheromone on the ground for foraging is the inspiration for the Ant colony optimization (ACO) algorithm. For the better image edge detection ACO is used in number plate recognition .This approach is able to establish a pheromone matrix that represents the information presented at each pixel position of the image and according to the movements of the number of ants which are dispatched to move on the image. The local variation of the images intensity values are driven by the movements of the ants. This system classification neural network in pattern recognition, artificial neural network (ANN), eventually this gives the number plate area extracted from the image with improved accuracy. Finally a character recognition model is used to give out the final vehicle license number.