Vehicle Identification and Tracking System based on Shape Parameters Analysis

In the existing work, it is observed that the algorithms are language dependent i.e. characters and numbers should be in local language and if the language is changed, then the character identification algorithm will have to be changed each time the language changes. This kind of problem has been targeted in the proposed work by using the statistical features of the text and can be fine-tuned easily just by replacing the statistical features data base of the particular language characters. Though, it is observed that the vehicle is not only recognized by its number plate but the model of the motor vehicle also possesses many information so that it could be identified from the distance as the number plate can only be extracted once the camera is focussed on it. But at the first sight, the vehicle is identified by it manifestation i.e. its shape, model, colour or style of the body. Therefore, in order to identify a vehicle on road, prima facie, its data in terms of its body or appearance is an important source of information that must be retrieved or extracted from its image. KeywordsHistogram of oriented gradients, Automatic Number Plate Recognition, binarizing