Vehicle license plate detection and recognition using non-blind image de-blurring algorithm

Abstract

This paper proposes the method of vehicle license plate recognition, which is essential in the field of intelligent transportation system. The purpose of the study is to present a simple and effective vehicle license plate detection and recognition using non-bling image de-blurring algorithm. The sharpness of the edges in an image is restored by the prior information on images. The blue kernel is free of noise while using the non-blind image de-blurring algorithm. This non-blind image de-blurring (NBID) algorithm is involved in the process of removing the optimization difficulties with respect to unknown image and unknown blur. Estimation is carried out for the length of the motion kernel with Radon transform in Fourier domain. The proposed algorithm was tested on different vehicle images and achieved satisfactory results in license plate detection. The experimental results deal with the efficiency and robustness of the proposed algorithm in both synthesized and real images.