Vehicle License Plate Recognition With Deep Learning

View Sample PDF

Author(s): Chi-Hsuan Huang (National Taiwan University, Taiwan), Yu Sun (National Taiwan University, Taiwan)and Chiou-Shana Fuh (National Taiwan University, Taiwan)
Copyright: 2022
Pages: 59
Source title: Technologies to Advance Automation in Forensic Science and Criminal Investigation
Source Author(s)/Editor(s): Chung-Hao Chen (Old Dominion University, USA), Wen-Chao Yang (National Central Police University, Taiwan)and Lijia Chen (Henan University, China)
DOI: 10.4018/978-1-7998-8386-9.ch009

Keywords: Criminal Science and Forensics / Digital Crime & Forensics / Information Science Reference / Security & Forensics

Purchase

View Vehicle License Plate Recognition With Deep Learning on the publisher's website for pricing and purchasing information.

Abstract

In this chapter, an AI (artificial intelligence) solution for LPR (license plate recognition) on moving vehicles is proposed. The license plates in images captured with cameras on moving vehicles have unpredictable distortion and various illumination which make traditional machine vision algorithms unable to recognize the numbers correctly. Therefore, deep learning is leveraged to recognize license plate in such challenging conditions for better recognition accuracy. Additionally, lightweight neural networks are chosen since the power supply of scooter is quite limited. A two-stage method is presented to recognize license plate. First, the license plates in captured images are detected using CNN (convolutional neural network) model and the rotation of the detected license plates are corrected. Subsequently, the characters are recognized as upper-case format (A-Z) and digits (0-9) with second CNN model. Experimental results show that the system achieves 95.7% precision and 95% recall at high speed during the daytime.

The IRMA Community

Research IRM

Vehicle License Plate Recognition With Deep Learning

Purchase

Abstract

Related Content

IRMA Sponsors