License Plate Detection Based on Cascaded Three Distinct Features

Abstract

As license plate detection is one of the key modules in license plate recognition system, various kinds of license plate detection methods have been proposed. In this paper, we propose an effective license plate detection method based on contour and internal features. We utilize the facts that the contour of a license plate is rectangle, intensity distribution of a license plate region is bimodal, and there exists specific number of characters in a license plate. Three classifiers are then designed based on the proposed features, and finally arranged in cascade manner. Experimental results show high accuracy of the proposed method.