License Plate Research Articles

Number Plate Detection Using CNN

This paper presents a deep learning- based approach for automatic number plate detection using Convolutional Neural Networks (CNN). The system focuses on accurately identifying and localizing vehicle license plates in real-time from images captured under various conditions. A custom CNN model is trained on a diverse dataset to extract spatial features and detect number plates effectively. The proposed method demonstrates high accuracy, robustness, and practical applicability in traffic monitoring and intelligent transportation systems.

Efficient Real-Time License Plate Recognition for Moroccan Vehicles Using Advanced Machine Learning Techniques

This study presents a machine-learning model designed for real-time recognition of Moroccan license plate characters, with a strong emphasis on practical deployment and real-world applicability. Addressing the structural and environmental challenges unique to Moroccan plates, the model utilizes a custom dataset of approximately 4,000 images, capturing essential characteristics for accurate identification. The approach integrates a Haar Cascade classifier for plate detection with a K-Nearest Neighbors (KNN) classifier for character recognition, ensuring robust performance across varying lighting conditions and partial occlusions. Evaluated using the accuracy metric, the model consistently achieved a high accuracy of 99%. A key focus of this work is the step-by-step implementation, making it well-suited for resource-limited environments such as traffic management, security, and law enforcement. This study provides a practical and efficient solution that will contribute to the advancement of license plate recognition technologies in regions facing similar challenges, offering a scalable framework for broader applications.

Opto-Electrical Decoupling of Phototransistors via Light-Induced Ferroelectric Depolarization for In-Sensor Computing.

Highly sensitive sensors are critical for in-sensor computing, an ultrafast and low-power machine vision technology. However, capturing sharp images without motion blur in low-light and high-speed situations remains challenging due to weak photoresponse. Here, we present a heterostructure ferroelectric phototransistor leveraging opto-electrical decoupling for fast perception and in-sensor computing. The channel is preprogrammed to a low-resistance state via ferroelectric polarization, while light modulates the drain current through light-induced ferroelectric depolarization. This mechanism enables a record-high MoTe2-based photoresponsivity of 3.05×104 A/W by optimizing the balance between depolarization and screening fields. The sensors can perceive light pulses as short as 200 μs, achieving an operating frequency of 5 kHz and an energy consumption of 74 fJ. Utilizing a light-programmable neutral point, a 3 × 3 sensor array was developed as the optical kernel for scene-specific in-sensor computing, achieving a license plate recognition accuracy of 92.4% with significantly reduced motion blur. These results demonstrate the potential of this technology for high-speed, low-light machine vision applications.

Real-Time License Plate Recognition and Alert System

ABSTRACT: The advent of deep learning has revolutionized computer vision, enabling real-time analysis crucial for traffic management and vehicle identification. This research introduces a system combining vehicle make and model detection with Automatic Number Plate Recognition (ANPR), achieving a groundbreaking 97.5% accuracy rate. Unlike traditional methods, which focus on either make and model detection or ANPR independently, this study integrates both aspects into a single, cohesive system, providing a more holistic and efficient solution for vehicle identification, ensuring robust performance even in adverse weather conditions. The paper explores the use of deep learning techniques, including OpenCV, in combination with Python programming language. Leveraging MobileNet-V2 and YOLOx (You Only Look Once) for vehicle identification, and YOLOv4-tiny, Paddle OCR (optical character recognition), and SVTR-tiny for ANPR, the system was rigorously tested at Firat University’s entrance with a thousand images captured under various conditions such as fog, rain, and low light. The system’s exceptional success rate in these tests highlights its robustness and practical applicability. Additionally, experiments evaluate the system’s accuracy and effectiveness, using Gradient-weighted Class Activation Mapping (GradCam) technology to gain insights into neural networks’ decision-making processes and identify areas for improvement, particularly in misclassifications. The implications of this research for computer vision are significant, paving the way for advanced applications in autonomous driving, traffic management, stolen vehicles, and security surveillance. Achieving real-time, high-accuracy vehicle identification, the integrated Vehicle Make and Model Recognition (VMMR) and ANPR system sets a new standard for future research in the field. INDEX TERMS: License plate recognition, deep learning, Open CV, human machine interface

Enhancing Parking Systems with QR Code-Integrated Automatic License Plate Recognition through Convolutional Neural Networks

This abstract describes the development and evaluation of an Automatic License Plate Recognition (ALPR) system designed to simplify the process of parking ticket generation. The traditional paradigm of manual entry of license plate information by parking personnel for exiting vehicles is replaced by the automated system proposed in this study. The system integrates a YOLO (You Only Look Once) model for the automatic recognition of license plates in vehicle images. After this initial identification, a series of pre-processing and image segmentation techniques are applied to isolate and recognize the individual digits within the license plate. A ResNet model is then used to classify the license plates. The research focuses specifically on Malaysian license plates. The experimental results show that the YOLO model recognizes license plates robustly and accurately and has a high degree of reliability. However, when validating the data set, the ResNet model achieves an accuracy of around 80 %. The study points out inherent challenges, including potential errors in segmentation, problems with non-standardized or damaged tags, and the presence of digits that may have visual similarities. In summary, while the YOLO model is reliable in recognizing license plates, the classification accuracy of the ResNet model can be further improved. Overcoming challenges such as segmentation noise and variations in license plate conditions could further optimize the overall performance of the system.

Enhancing Road Safety MI-Powered Helmet and License Plate Detection

Abstract: Individuals frequently disregard how important it is to wear helmets, which causes tragic accidents. A helmet reduces your risk of getting a serious brain injury and dying by deflecting most of the impact energy that would otherwise hit your head and brain during a tumble or collisions. In India, it is against the law to operate a motorbike or scooter without a helmet, which has increased fatalities as well as crashes. The existing system mostly relies on surveillance footage for keeping up with traffic violations, necessitating a close-up of the license plate by traffic police in the case that the motorcyclist lacks a helmet. Yet, this necessitates a substantial amount of personnel and time considering the high frequency of traffic violations and the rising everyday use of motorcycles. Imagine if there was an algorithm that monitored traffic infractions, such driving a motorbike with no a helmet, and, if any were identified, generate the license plate of the vehicle that committed the violation. Helmet and license plate is detected using a neural network is proposed in this paper. There will be two phases. Initially, we check to see if the riders are wearing helmets. If not, a second step is used to find their license plate. To identify unauthorized vehicles, we also look for license plates on passing vehicles. Keywords: Convolutional Neural Network, Helmet Detection, Machine Learning, Yolov8

Benchmarking Probabilistic Deep Learning Methods for License Plate Recognition

Abstract: This study benchmarks probabilistic deep learning methods for license plate recognition (LPR), focusing on enhancing accuracy and reliability under real-world conditions. Utilizing a dataset of license plate images, the approach includes comprehensive preprocessing steps such as resizing, normalization, augmentation, and super-resolution to handle low-quality inputs. The dataset is split into training, validation, and testing subsets, with the test set emphasizing out-of-distribution (OOD) scenarios. The system employs convolutional neural networks (CNNs), probabilistic models like SR2 methods to estimate prediction uncertainty. A multi-task learning model is introduced to simultaneously perform LPR and image super-resolution, leveraging shared features for improved performance. Evaluation metrics include accuracy, precision, recall, F1-score, mean squared error, and novel uncertainty-based measures such as prediction uncertainty and error detection rate. Results demonstrate a 109% accuracy improvement with the multi-task model and a 29% increase in error detection via uncertainty quantification, highlighting the system's robustness and practical value in uncertain environments. Keywords: License Plate Recognition (LPR), Probabilistic Deep Learning, Uncertainty Quantification, Multi-Task Learning, Super-Resolution, Out-of-Distribution (OOD), Bayesian Neural Networks

AI-Powered Smart Traffic Management System

ABSTRACT Urban traffic congestion and rule violations remain significant challenges for modern cities. This paper proposes an AI-Powered Smart Traffic Management System that integrates real-time video analysis, intelligent violation detection, license plate recognition, and automated challan issuance through Telegram. The system uses computer vision and machine learning algorithms to identify traffic violations such as over-speeding, helmet non-compliance, seatbelt negligence, and red-light jumping. License Plate Recognition (LPR) is used to identify the vehicle, and an automated challan is sent to the violator. A Telegram bot is integrated for real-time notification, and a back-end database maintains logs. Experimental results demonstrate high accuracy and real-time performance, highlighting the potential for deployment in smart cities to improve road safety and reduce human dependency in traffic monitoring.Keywords - Deep Learning, Natural Language Processing, Symptomatic Neural Network, machine learning, Mental Health Chat bot, Text. and Speech Processing, Depression Detection Keywords—Smart traffic, AI, deep learning, license plate recognition, traffic violation, e-challan, Telegram bot, computer vision.

Smart Traffic Management and Detection System Using AI and Computer Vision

ABSTRACT The Smart Traffic Management and Violation Detection System integrates AI and Computer Vision to optimize traffic control and detect violations. By analyzing real-time video feeds from surveillance cameras, it adjusts traffic signal timings based on vehicle density, reducing congestion and delays. The system automatically detects violations such as red light jumping, helmetless riding, over speeding, and wrong-lane driving using object detection and OCR techniques. Violators’ license plates are recorded, and penalty notices are generated automatically. This solution enhances road safety, improves compliance, supports law enforcement, and provides valuable data for future urban planning and infrastructure development. Keywords: Smart Traffic Management, Traffic Violation Detection, Artificial Intelligence (AI), Computer Vision, Real-Time Video Analysis.

Automated Vehicle License Plate Detection System Using MATLAB

Abstract - With rising industrialization, the demand for vehicles has risen proportionally, which has introduced various challenges in vehicle management and traffic control. An efficient system for vehicle monitoring and safety has become essential to resolve these issues. Hence, an automated license plate detection system using OCR plays a vital role in improving security and accurate vehicle identification. The proposed system utilizes advanced image processing techniques to detect vehicle license plates and extract plate information efficiently. This system ensures fast and accurate identification, capturing vehicle details as they enter a monitored area. With the growing demand for smart and automated solutions, the implementation of such technology can significantly enhance traffic management, security enforcement, and transportation efficiency. In this paper, we focus on various aspects of image processing implemented in our system to simplify processing. This system efficiently detects vehicle license plates and extracts the plate information. The system is based on MATLAB software to improve detection accuracy. Our proposed system covers applications related to security and law enforcement. It is particularly useful at places where vehicle monitoring is essential. By using cameras and software, the system ensures accurate vehicle detection and enhances security. Key words- Industrialization, Image Processing, OCR, MATLAB, Law Enforcement.

Enhancing Motorcycle Safety and Security Using Biometric Verification and Helmet Detection Through AI and ML Integration

The automotive industry, like many others, has had to adapt to newer technologies, including biometrics, AI, and Machine Learning (ML) over the years. In this chapter, a sophisticated technique directed on improving vehicle security through biometric verification and helmet detection is described using TensorFlow and a custom-built Arduino-based system. A notable factor contributing to fatalities in these motorcycle accidents is lack of helmet use among riders. Ensuring that riders put on helmets before they get on their motorcycles is done through monitoring with CCTV cameras or by manually supervising traffic intersections by police. These methods however entail high levels of manual labor. Through these approaches, some motorcycle riders without helmets have been tracked, and their license plates captured via CCTV cameras. In this methodology, the algorithm first identifies moving objects as motorcycles or non-motorcycles. Moreover, the system assesses whether classed helmeted riders are wearing helmets. If not, the device applies an OCR technique to capture the license plate

Prediction of electric vehicles CO2 emission trajectory and peak time series in China.

Prediction of electric vehicles CO2 emission trajectory and peak time series in China.

A heterogeneous two-layer graph convolution model for turning traffic prediction with missing data

Turning traffic prediction at urban intersections is very important for the dynamic optimization of traffic management strategies, but accuracy is often influenced by missing data. We propose a novel approach with a Heterogeneous Two-Layer Graph Convolution (HTLGC) model to enhance prediction accuracy while addressing missing data challenges. We construct the urban road network as a heterogeneous two-layer spatial graph, with intersection nodes in the upper layer and turning nodes in the lower layer. To address the missing values, we introduce a feature propagation algorithm. The spatial module equipped with the attention mechanism is used to capture these two distinct levels of spatial information. Moreover, the temporal pattern attention module is applied to more effectively mine features over time. Experiments using license plate recognition data from Xi'an, China, demonstrate that our HTLGC model outperforms baseline algorithms under various missing data rates.

Smart Parking Your Parking Assistant

Abstract—Smart Park It is a software solution designed to enhance the parking experience in locations with a high volume of vehicles, such as airports, shopping malls, and other busy venues. Upon entering a mall, drivers must find a parking space for their motorcycles or cars, often relying on security personnel to assist in locating an available slot. Retrieving a vehicle from a crowded parking area can be challenging, as it requires remembering the exact location of the parked vehicle. While this task may be manageable in smaller facilities, it becomes increasingly difficult in larger environments like international airports. To address this issue, we propose a software application that utilizes AI technology. Upon arrival at the parking facility, the vehicle's license plate will be scanned by AI cameras at the entrance, and a ticket featuring slot address and a QR code will be issued. Scanning this QR code will provide directions to the designated parking space, facilitating both the parking and retrieval processes. The system will require two cameras: one at the entrance to capture the license plate and another within the parking area to identify available slots. Introduction

Smart Traffic Surveillance: Helmet Detection and License Plate Recognition Using Deep Learning

Abstract— Motorcyclists are particularly vulnerable to road accidents, often resulting in severe injuries or fatalities. Helmets are a proven safety measure that significantly reduces the risk of head injuries; however, many riders fail to comply with helmet laws, making enforcement a challenge for traffic authorities. This paper presents an advanced, automated traffic violation detection system that integrates real-time helmet detection and automatic license plate recognition to enhance road safety and streamline enforcement. Our system employs an optimized YOLOv3 model to efficiently detect motorcycles and determine whether riders are wearing helmets. Unlike conventional implementations, we enhance detection accuracy by fine-hyperparameter tuning YOLOv3 on a diverse dataset that includes various helmet types, different lighting conditions, and occlusions. To address challenges such as low-light environments, we incorporate preprocessing techniques, including contrast enhancement and adaptive thresholding, improving detection performance under suboptimal conditions. For license plate recognition, we utilize EasyOCR, further improved through custom preprocessing steps such as noise reduction and edge enhancement and thresholding enabling better recognition of partially occluded or low-quality license plates. Upon detecting a violation, the system automatically extracts and logs the license plate details into a structured database, facilitating streamlined enforcement and legal action. Our experimental results demonstrate increased accuracy while reducing inference time compared to existing methods, making this system a scalable and deployable solution for real-time traffic monitoring. By automating violation detection and reporting, the proposed approach reduces the burden on law enforcement while encouraging greater compliance with helmet laws, ultimately contributing to safer roads. Keywords— YOLO, EasyOCR, preprocessing techniques, traffic regulations and real-time monitoring.

Automatic Number Plate Recognition Using YOLOv8 Model

Automatic Number Plate Recognition (ANPR) systems have become a critical tool in various sectors, including traffic management, law enforcement, and tolling systems. This paper presents an in-depth exploration of an advanced ANPR framework that leverages cutting-edge image processing methodologies and machine learning models to deliver exceptional accuracy in license plate detection and recognition. The system follows a multi-phase approach encompassing image capture, preprocessing, plate localization, character segmentation, and optical character recognition (OCR). Notably, the integration of YOLOv8, a state-of-the-art deep learning model for object detection, significantly enhances the feature extraction and classification process, boosting the system's performance across diverse environmental challenges. The proposed approach achieves a recognition accuracy exceeding 95%, highlighting its potential for deployment in real-world scenarios. Additionally, the paper addresses various challenges encountered in ANPR systems, such as variations in license plate formats, fluctuating lighting conditions, and partial occlusions, and proposes future research directions aimed at further improving robustness and operational efficiency.

Advanced AI-Based License Plate Recognition and Pollution Com-pliance Monitoring for Motorbikes Using YOLOv8 and LLaMA OCR

Motorbike traffic has significantly increased in urban areas, posing challenges for traffic management and environmental compliance. Existing License Plate Recognition (LPR) systems struggle with rec- ognizing non-standard motorbike plates, especially under adverse conditions such as low-light envi- ronments, occlusions, and distorted fonts. Additionally, real-time enforcement of pollution compliance remains underexplored. This study presents an advanced AI-based approach integrating YOLOv8 for high-precision license plate detection and LLaMA OCR, a transformer-based model, for robust char- acter recognition in challenging conditions. Furthermore, a Real-Time Compliance Monitoring Mod- ule (RT-CMM) is introduced to verify vehicle registration and pollution compliance through government databases. Experimental results on diverse datasets, including Indian motorbike plates, demonstrate an impressive 99.7% detection accuracy and 97.8% compliance verification success rate. The pro- posed system outperforms conventional methods and provides a scalable solution for smart traffic mon- itoring, regulation enforcement, and environmental protection in urban settings

A Deep Learning Framework for Accurate Number Plate Recognition using OWL-V2 and PaddleOCR

Recognizing number plates holds significant importance in various applications such as law enforcement, traffic management, toll collection, parking management, and security. Manual recognition is challenging one and is prone to errors. Automation of number plate recognition enables faster inference and timely response from the police personnel. Artificial intelligence techniques using deep learning models are useful in this regard. Typically, Automated Number Plate Recognition (ANPR) technology to identify and process vehicle license plates effectively Such as Image Capture, Image Processing, Character Recognition, Alerts or Actions. This paper presents a vision-language model-based approach for automatic detection. Number Plate Recognition (NPR) is a technique that identifies alphanumeric characters from license plates and converts them into text format. In this work, different deep learning models were utilized for the detection and recognition of number plates. The model was tested on a car image, processed and number plate datasets. For this recognition task, zero-shot models such as OWLVIT and Grounding DINO, were employed. Additionally, techniques like PaddleOCR were integrated. The proposed tests demonstrated that the system can accurately detect number plates with an impressive accuracy of 92.91%, even under challenging conditions.

Automatic License Plate Recognition

Automatic license plate recognition using Image processing offer students a unique opportunity to gain hands-on experience in designing and optimizing the Embedded system. This typically involve working with industry professionals on actual projects, providing interns with valuable exposure to real-world challenges and best practices in the field of Embedded system. During the project, participants are often tasked with designing using specialized software tools like MATLAB. In recent times, the number of vehicles on road has exponentially risen due to which traffic congestion and violations are a menace on roads. Automatic License Plate Recognition (ALPR) system can be used to automate the process of traffic management thereby easing out the flow of traffic and strengthening the access control systems. In this paper, we compare the efficiency achieved by morphological processing and edge processing algorithms. A detailed analysis and optimization of neutral network parameters such as regularization parameter, number of hidden layer units and number of iterations is done. The system utilizes image processing techniques and machine learning algorithms running on MATLAB and Arduino to obtain the results with an efficiency of 97%. The system was tested on a set of static images as well as in a dynamic environment. The execution time of the system on a dynamic environment and the license plate of the vehicle can be identified as well. In addition to technical skills, interns gain exposure to the latest industry trends and technologies, preparing them for future roles in the Embedded of electronics industries. The mentorship and networking opportunities available during the project can lead to valuable connections and potential job offers after graduation. Keywords: License Plate Recognition, Morphological Operations, Character Segmentation, Template Matching, Sobel Filter, AVR Microcontroller

Exploring outdoor servicescapes in the sharing economy: a thematic analysis of motorcycle ride-hailing and early adults’ experiences

PurposeThis study aims to explore the dimensions of the outdoor servicescape of motorcycle ride-hailing transportation, focusing on the experiences of early adults as defined by Levinson’s adult development theory.Design/methodology/approachUsing a qualitative thematic analysis, this study investigates the dimensions of motorcycle ride-hailing servicescapes, centering on early adults’ (aged 23–40) perspectives. This demographic’s transitional life stage, coupled with their reliance on digital platforms and shared resources, positions them as key stakeholders in the sharing economy.FindingsThe study identifies key servicescape dimensions: spatial layout and functionality (motorcycle type/size, seat comfort, vehicle readiness, engine condition, safety equipment and passenger helmets); signs and symbols (license plates and drivers’ attributes); interaction and behavior (communication, route proficiency, driving conduct and general etiquette); and online servicescape [in-app communication, map and route features, payment methods and user interface (UI) or user experience (UX)]. These findings highlight the interplay between the physical servicescape of the motorcycle, the social servicescape of the driver and the online servicescape of the digital application, all within the resource-sharing dynamics that define the motorcycle ride-hailing experience in an outdoor environment.Research limitations/implicationsThis study focuses exclusively on early adults, limiting generalizability to other age groups, such as teenagers or older adults, who may also be frequent users of ride-hailing services.Practical implicationsTheoretically, this study extends servicescape research to outdoor, utilitarian contexts within the sharing economy, addressing gaps in literature focused on static, indoor or hedonic environments. Practically, the findings offer actionable implications for enhancing service quality in regions like Southeast Asia and Africa, where motorcycle ride-hailing services are prevalent. Companies should implement strict vehicle maintenance, provide comprehensive driver training and optimize app design with features like seamless UI/UX, accurate maps, digital payment options and transparent pricing to improve user satisfaction and trust.Originality/valueTo the best of the authors’ knowledge, this study is among the first to examine outdoor servicescapes in the context of the sharing economy, offering novel insights into motorcycle ride-hailing services and their relevance to early adults.