DefinePK

Abstract

The target of this paper is to build up a well-organized algorithm hinged on Automatic Vehicle Registration Plate Recognition system. It is the mining of License Plate data from vehicle pictures or videos, which accommodates three parts such as License Plate Localization, Character Segmentation and Character Recognition. In this paper, we have projected an efficient License Plate Localization algorithm subjected to Adaptive Neuro Fuzzy Inference System. This proposal involves a mixture phases of pre-processing such as Gray Scale conversion, Sobel edge map detection and morphological image binarization. Followed by the Maximally Stable External Region text region detection, geometrical and texture features are drawn out to be granted as input to ANFIS for the training purpose for localizing the License Plates. The proposed method is deliberated to achieve localization of any category of License Plates covering at all categories of environmental surroundings. In this efficient method of License Plate Localization, the empirical results express reasonably forceful localization rate as 93.61%. The proposed system provides a rapid sprouting technology and a fundamental constituent for the Intelligent Transport Systems.