In last decade, there has been explosive growth in vehicular sector and the number of vehicles flying on the road in all the parts of the country. For the better management of vehicular traffic, it is necessary to keep track of vehicles on the basis of their number plates. The proposed project suggests an automated way of parking toll collection based on the number plates of the vehicle and the time for which the vehicle is parked in the parking lot. This work deals with the problem from field of artificial intelligence, computer vision (image processing) and neural networks in the construction of an Automatic Number Plate Recognition System (ANPR).This problem includes mathematical principles and algorithms, which ensures a various processes to carry out the steps for the product. The acquisition of the image takes place with any camera with capability of capturing image with good quality. The emphasis of this paper is on the localization of number plate using contours tracing technique along with edge detection and sharpening of edge using Canny’s edge detection algorithm. Moreover, we focus on this paper a new algorithm based on artificial neural network (ANN) is used for recognition of number plate characters. This paper makes use of various algorithms in each category from number plate detection to actual recognition of characters which enhances the performance of the system up to the maximum extent possible with less efforts and use of computational resources.