Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications

Road traffic emissions are still a significant contributor to air pollution, which causes adverse health effects. Remote emission sensing (RES) is a state-of-the-art technique that continuously monitors the emissions of thousands of vehicles in traffic. Automatic number plate recognition (ANPR) syst...

Full description

Saved in:
Bibliographic Details
Main Authors: Hafiz Hashim Imtiaz, Paul Schaffer, Paul Hesse, Martin Kupper, Alexander Bergmann
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/25/11/3499
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849722441636511744
author Hafiz Hashim Imtiaz
Paul Schaffer
Paul Hesse
Martin Kupper
Alexander Bergmann
author_facet Hafiz Hashim Imtiaz
Paul Schaffer
Paul Hesse
Martin Kupper
Alexander Bergmann
author_sort Hafiz Hashim Imtiaz
collection DOAJ
description Road traffic emissions are still a significant contributor to air pollution, which causes adverse health effects. Remote emission sensing (RES) is a state-of-the-art technique that continuously monitors the emissions of thousands of vehicles in traffic. Automatic number plate recognition (ANPR) systems are an essential part of RES systems to identify the registered owners of high-emitting vehicles. Recognizing number plates on L-vehicles (two-wheelers) with a standard ANPR system is challenging due to differences in size and placement across various categories. No ANPR system is designed explicitly for Category L vehicles, especially mopeds. In this work, we present an automatic number plate detection and recognition system for Category L vehicles (L-ANPR) specially developed to recognize L-vehicle number plates of various sizes and colors from different categories and countries. The cost-effective and energy efficient L-ANPR system was implemented on roads during remote emission measurement campaigns in multiple European cities and tested with hundreds of vehicles. The L-ANPR system recognizes Category L vehicles by calculating the size of each passing vehicle using photoelectric sensors. It can then trigger the L-ANPR detection system, which begins detecting license plates and recognizing license plate numbers with the L-ANPR recognizing system. The L-ANPR system’s license plate detection model is trained using thousands of images of license plates from various types of Category L vehicles across different countries, and the overall detection accuracy with test images exceeded 90%. The L-ANPR system’s character recognition is designed to identify large characters on standard number plates as well as smaller characters in various colors on small, moped license plates, achieving a recognition accuracy surpassing 70%. The reasons for false recognitions are identified and the solutions are discussed in detail.
format Article
id doaj-art-25e4f0c0aa664af98f51ee3fb6c8f5bf
institution DOAJ
issn 1424-8220
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-25e4f0c0aa664af98f51ee3fb6c8f5bf2025-08-20T03:11:20ZengMDPI AGSensors1424-82202025-05-012511349910.3390/s25113499Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing ApplicationsHafiz Hashim Imtiaz0Paul Schaffer1Paul Hesse2Martin Kupper3Alexander Bergmann4Institute of Electrical Measurement and Sensor Systems, Graz University of Technology, 8010 Graz, AustriaInstitute of Electrical Measurement and Sensor Systems, Graz University of Technology, 8010 Graz, AustriaInstitute of Electrical Measurement and Sensor Systems, Graz University of Technology, 8010 Graz, AustriaInstitute of Electrical Measurement and Sensor Systems, Graz University of Technology, 8010 Graz, AustriaInstitute of Electrical Measurement and Sensor Systems, Graz University of Technology, 8010 Graz, AustriaRoad traffic emissions are still a significant contributor to air pollution, which causes adverse health effects. Remote emission sensing (RES) is a state-of-the-art technique that continuously monitors the emissions of thousands of vehicles in traffic. Automatic number plate recognition (ANPR) systems are an essential part of RES systems to identify the registered owners of high-emitting vehicles. Recognizing number plates on L-vehicles (two-wheelers) with a standard ANPR system is challenging due to differences in size and placement across various categories. No ANPR system is designed explicitly for Category L vehicles, especially mopeds. In this work, we present an automatic number plate detection and recognition system for Category L vehicles (L-ANPR) specially developed to recognize L-vehicle number plates of various sizes and colors from different categories and countries. The cost-effective and energy efficient L-ANPR system was implemented on roads during remote emission measurement campaigns in multiple European cities and tested with hundreds of vehicles. The L-ANPR system recognizes Category L vehicles by calculating the size of each passing vehicle using photoelectric sensors. It can then trigger the L-ANPR detection system, which begins detecting license plates and recognizing license plate numbers with the L-ANPR recognizing system. The L-ANPR system’s license plate detection model is trained using thousands of images of license plates from various types of Category L vehicles across different countries, and the overall detection accuracy with test images exceeded 90%. The L-ANPR system’s character recognition is designed to identify large characters on standard number plates as well as smaller characters in various colors on small, moped license plates, achieving a recognition accuracy surpassing 70%. The reasons for false recognitions are identified and the solutions are discussed in detail.https://www.mdpi.com/1424-8220/25/11/3499automatic number plate detection and recognitionmotorcycle number plate detectionimage processingcomputer visionconvolutional neural networksremote emission sensing
spellingShingle Hafiz Hashim Imtiaz
Paul Schaffer
Paul Hesse
Martin Kupper
Alexander Bergmann
Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications
Sensors
automatic number plate detection and recognition
motorcycle number plate detection
image processing
computer vision
convolutional neural networks
remote emission sensing
title Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications
title_full Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications
title_fullStr Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications
title_full_unstemmed Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications
title_short Automatic Number Plate Detection and Recognition System for Small-Sized Number Plates of Category L-Vehicles for Remote Emission Sensing Applications
title_sort automatic number plate detection and recognition system for small sized number plates of category l vehicles for remote emission sensing applications
topic automatic number plate detection and recognition
motorcycle number plate detection
image processing
computer vision
convolutional neural networks
remote emission sensing
url https://www.mdpi.com/1424-8220/25/11/3499
work_keys_str_mv AT hafizhashimimtiaz automaticnumberplatedetectionandrecognitionsystemforsmallsizednumberplatesofcategorylvehiclesforremoteemissionsensingapplications
AT paulschaffer automaticnumberplatedetectionandrecognitionsystemforsmallsizednumberplatesofcategorylvehiclesforremoteemissionsensingapplications
AT paulhesse automaticnumberplatedetectionandrecognitionsystemforsmallsizednumberplatesofcategorylvehiclesforremoteemissionsensingapplications
AT martinkupper automaticnumberplatedetectionandrecognitionsystemforsmallsizednumberplatesofcategorylvehiclesforremoteemissionsensingapplications
AT alexanderbergmann automaticnumberplatedetectionandrecognitionsystemforsmallsizednumberplatesofcategorylvehiclesforremoteemissionsensingapplications