Approach of Car License Plate Location based on Improved Sobel Operator with the Combination of Mathematical Morphology

Abstract

Plate license locating is the key process in vehicle license plate recognition for automated transport system. It will affect the accuracy of recognition directly. License plate area contains rich edge and texture information. In this paper, improved Sobel operator is used to detecting the edge of the car image. This paper presents a new approach of extracting car license plate from car images based on morphology. Based on the information of width and height of car license plate, and space between characters of the car license plate, we design a new structural elements for morphological operations used to the edge image of car license plate, which can obtain several candidate areas. Finally, the criterions basing on the ratio of plate area, ratio of height to width and projections are used in comprehensive analysis,and the real license plate area can be located accurately. The improved Sobel operator method is helpful to reveal distinct characteristics for some edge features, efficiently with high speed processing. Compared with the classical edge detection operators edge detection algorithm, the improved algorithm is more advantage, the quality and speed of the picture edge extraction significantly improved. Experimental results show that this method is simple, accurate.It can satisfy the real time need of license location demand, and can make the preparation for the following character identification.