Automatic License Plate Recognition Using Improved Deep Learning

Motamed, Sara

doi:10.22052/scj.2024.253175.1163

Automatic License Plate Recognition Using Improved Deep Learning

Articles in Press

Document Type : Original Article

Author

Sara Motamed

Department of Computer Engineering, Fouman and Shaft Branch, Islamic Azad University, Fouman, Iran

10.22052/scj.2024.253175.1163

Abstract

According to today's statistics, more than half a billion vehicles are moving in the world and inspection and monitoring is one of the basic needs of any traffic system. All cars have an identification number or the same license plate as their primary ID, which today is one of the most suitable vehicle authentication tools. In this paper, deep learning methods are used to detect license plates. The proposed method includes two stages of highlighting the license plate and reading the ID. In this regard, the combination of deep neural networks and competitive generative network (GAN) in encoding-coder network has been used to highlight. The proposed models have been evaluated on FZU Cars and Stanford Cars datasets and the results of this study have been discussed. By examining the results, we find that the proposed model has reached an accuracy of nearly 98% in both datasets.

Keywords