期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：4
页码：6739
DOI：10.15680/IJIRCCE.2017.0504024
出版社：S&S Publications
摘要：Abstract: Recently, the study on License number plate recognition is one of the most inspiring andambitious task. Since, it is applied to Border crossing vehicle, toll-collection at highways, traffic management, parkingmanagement at variant locations etc. In addition to, when any vehicle is missing, the license plate assists to find themissing vehicles and also location of the accidents. The over-speed vehicle captured in the surveillance camera is notvisible to human because of its low-resolution. Thus, the issue, license plate image blurring under fast motion isconvoluted in linear with modeled angle and length should be studied. In this paper, we have proposed a sparserepresentationmodel that detects the number plate in fast motion system. To do so, the sparse coefficients arecalculated for recovered image when kernel angle corresponds to motion angle. Then, the kernel length is estimatedwith radon transform in Fourier domain. Atlast, plate number recognition process is then applied for blur removal. Theplate number recognition process is done in three steps, namely, Character segmentation, Optical character recognitionand template matching. Experimental analysis is carried out using MATLAB, programming language. A performanceresult is achieved in terms of accurate detection with minimized time duration.
