Please use this identifier to cite or link to this item:
Title: Automatic car plate recognition using artificial intelligence
Authors: Zhu, Ziwei
Keywords: Engineering::Electrical and electronic engineering
Issue Date: 2022
Publisher: Nanyang Technological University
Source: Zhu, Z. (2022). Automatic car plate recognition using artificial intelligence. Master's thesis, Nanyang Technological University, Singapore.
Abstract: As one of the critical technologies of the intelligent transportation system, the license plate recognition system has developed rapidly in recent years. In particular, new deep learning methods promote the update of license plate recognition algorithms. The license plate recognition algorithm can complete the recognition under complex conditions such as blurred images, uneven illumination, and tilted license plate. However, its vast computational load and slow recognition efficiency make it challenging to deploy the algorithm to the front of the device. For this reason, our primary work is as follows: (1)Design and implement an end-to-end real-time license plate recognition algorithm. We use the method of feature separation and feature recombination to design and implement a separate license plate detection network. It improves the speed of the license plate recognition algorithm effectively. (2)We improve the expression of the license plate shape as a five-dimensional vector and propose a rotating RoIAlign operation matching the five-dimensional vector. (3)We use the newest license plate dataset for model training and testing. Our model performs better and can be better applied to some scenes. Keywords: license plate recognition, end-to-end, Rotating RoIAlign.
Schools: School of Electrical and Electronic Engineering 
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
  Restricted Access
2.26 MBAdobe PDFView/Open

Page view(s)

Updated on Dec 8, 2023


Updated on Dec 8, 2023

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.