The goal of Automatic License Plate Recognition (ALPR) is localizing the license plate of a vehicle from an image and extracting text from it to recognize and track the vehicle. Each year, the amount of vehicles in Bangladesh is increasing at a significant rate. With the increasing number of vehicles, the intelligent transport system (ITS) has become essential. The automatic license plate recognition system (ALPRS) is a key part of ITS. The ALPRS can also help monitor traffic, surveillance of certain areas, crime investigations, etc. This paper has proposed an optimal end-to-end approach for the ALPR system for Bangladeshi vehicles by experimenting with the various deep neural network (DNN) models. These models have been trained and evaluated on our rich datasets of Bangladeshi vehicles and license plates. We have also introduced an algorithm that eliminates the need for the typical segmentation phase and generates properly formatted output efficiently. The final proposed system offers 99.37% accuracy in license plate localization and 96.31% accuracy in text recognition from the license plate (LP)s.
Automatic License Plate Recognition System for Bangladeshi Vehicles Using Deep Neural Network
Lecture Notes on Data Engineering and Communications Technologies
Proceedings of the International Conference on Big Data, IoT, and Machine Learning ; Chapter : 8 ; 91-102
2021-12-04
12 pages
Article/Chapter (Book)
Electronic Resource
English
Automatic license plate recognition
IEEE | 2004
|Automatic License Plate Recognition
Online Contents | 2004
|License Plate Recognition System Based on Neural Network
Springer Verlag | 2023
|European Patent Office | 2015
|