License Plate Recognition Research Articles

Calibration Method for Wide-Field Machine Vision System Based on Laser Projection Turntable

Calibration Method for Wide-Field Machine Vision System Based on Laser Projection Turntable

Examining Police Officers’ Perceptions of Automated License Plate Readers Before Technology Expansion

Automated license plate readers (ALPRs) are one of the most recent technological advancements that have rapidly diffused across U.S. law enforcement. A majority of the large police agencies utilize APRs, yet little empirical and evaluative research has been conducted on this technology. This study seeks to (a) synthesize what is known about ALPRs and (b) examine police officers’ perceptions of ALPRs before a major expansion of the technology in a single agency in the Mid-Atlantic region of the United States. Using an officer survey of 110 respondents, results found that those with prior experience using the technology, younger and more inexperienced officers, and those with stronger guardian orientations possessed more positive perceptions of the ALPR expansion. Prior experience using ALPRs also served as a protective factor by being associated with lower levels of skepticism toward the technology expansion; those officers with stronger warrior orientations had higher levels of skepticism. Practical considerations and directions for future research are discussed.

Beam profiles and radiation coherence at the output of solid-state and dye lasers with an intra-cavity immersion diffuser

Lasers on mixtures of solid particles and liquids have long been an attractive subject of research as they can serve as sources of low-coherent radiation and provide a good thermal management using a medium flow. One of such schemes is a laser with an intra-cavity cuvette-diffuser filled by a dense immersion mixture of activated solid micro-particles and a liquid (slurry), similar to the known Christiansen filter. The properties of perspective slurry lasers and features of beams formation in such media were studied in model experiments on bulk Nd:YVO4 and Nd:YAG active elements pumped by 808 nm diodes in quasi-continuous (1 ms, 10 Hz) regime, and on a Rhodamine 101 dye laser pumped by 20 ns, 532 nm light pulses. Laser media were tested in compact (7–50 mm) resonators using a diffuser on a mixture of non-doped LiF micro-particles and isobutanol. The laser spatial coherence was estimated by the contrast C of speckle patterns or by the visibility of Young interferograms γ. Near the threshold separate deformed Laguerre-Gaussian modes and concentric rings were observed at lasers’ output, indicating radiation coherence of C=0.5−0.7. At 100 W pumping Nd:YVO4 and Nd:YAG lasers with diffusers in plano-spherical (PS) resonators provided 1 ms pulses of 1–6 W multimode 1064 nm generation with smoothed beam profiles, divergence up to 0.11 rad and C=0.02−0.1. For a dye laser with a diffuser in PS resonators, 17 ns pulses of 627 nm multimode radiation with 1–3.5 mJ energy, beam divergence 60–280 mrad, γ=0.01−0.05 and up to 30% optical efficiency were obtained. Thus, an intra-cavity diffuser with an immersion mixture is a convenient and simple device to reduce the laser coherence. C and γ values, close to those for thermal light sources make it possible to use the developed lasers for speckles suppression at laser projection, in speed photography, etc. Promising solid-state and dye slurry lasers are discussed. The obtained results show the specific of beam formation in slurry media and give grounds to expect successful development and applications of low-coherent solid-state and dye slurry lasers operating in the visible and infra-red spectral regions.

Dual license plate recognition and visual features encoding for vehicle identification

This work presents an improved version of a new approach for vehicle identification, which comprises a dual identification system based on license plate recognition and visual encoding. To support this proposal, two new datasets have been created: UC3M-LP for license plate detection and character recognition and UC3M-VRI for vehicle re-identification. The main contributions of this research are the publication of the two open-source datasets and the validation of the dual approach for a reliable vehicle recognition. Precisely, the UC3M-LP dataset is unique, as it fills the gap of European license plates public datasets, becoming the largest of its kind and the first ever for Spanish plates. The proposed dual identification system provides a more robust solution, as it is less sensitive to the variability of image conditions. Performance has been evaluated both on public and the proposed datasets using a multi-network architecture and achieving remarkable results. This strategy opens up new research opportunities in the field of vehicle identification, and the generated datasets may serve as a benchmark for future research. The datasets are publicly available at https://github.com/ramajoballester/UC3M-LP and https://github.com/ramajoballester/UC3M-VRI.

License plate recognition system in unconstrained scenes via a new image correction scheme and improved CRNN

Automatic license plate recognition plays an important role in vehicle management. The performance of current license plate recognition systems is significantly affected by large-angle deflection in unconstrained scenes. For addressing this issue, this paper develops a license plate recognition system via a new image correction scheme and an improved CRNN (Convolutional Recurrent Neural Network). In which, the yolov5l is firstly introduced for license plate detection, and it is trained by transfer learning in order to address insufficient data. Then, a new license plate correction scheme and AFF-Net (Adaptive Fusion Feature Segmentation Network) is proposed and applied, which uses the segmentation result of license plate and its original image area for perspective transformation to improve the correction effect. Finally, the channel attention mechanism is added to the CRNN model for license plate recognition, so that a single grid cell of the feature network can obtain more spatial information. The recognition accuracy of the developed license plate recognition system is 98.8 % on CTPFSD (China Temporary Parking Fee System Data), which is much higher than current mainstream license plate recognition systems. More importantly, the developed license plate recognition system has been deployed in the cloud and applied in outdoor parking toll in practice.

CNN based System for Automatic Number Plate Recognition

This study presents a comprehensive approach to Automated Vehicle Number Plate Detection and Recognition, employing image processing and Convolutional Neural Networks (CNNs). The system encompasses two main stages: number plate detection and recognition. Utilizing a digital camera, the system employs image processing to segment the number plate region accurately. A super-resolution method is then applied via CNNs to enhance the image quality. Subsequently, a bounding box method isolates individual characters for precise recognition. In the recognition phase, CNNs extract features for effective classification. The study aims to advance automated vehicle identification systems for law enforcement and parking management applications, promising accurate and efficient number plate detection and recognition. The proposed work has also developed a user interface to ensure the successfulness of the objectives aimed.

Finding the Spot: IoT enabled Smart Parking Technologies for Occupancy Monitoring – A Comprehensive Review

Major cities in India have a significant number of vehicles, and the rate of ownership is increasing every day. However, the lack of proper parking infrastructure in these cities causes problems such as difficulty in finding parking spaces. According to the Urban Mobility Survey 2023 by Times Network, nearly 74% of vehicle owners in metropolitan cities struggle to find a parking slot. Various measures have been implemented to address this issue. One of the most promising measures is a smart parking management system. This system can use technologies like Radio Frequency Identification (RFID) and Automatic License Plate Recognition (ALPR) to make check-in and check-out easier. It can also include Wireless Sensor Networks (WSN), wired sensors, or visual occupancy detection to provide real-time occupancy status. The smart parking management system can offer useful services through mobile or web applications such as parking occupancy monitoring, reservation, payment gateway, occupancy prediction, automated check-in and check-out, and parking record management. The purpose of this review paper is to summarize the works undertaken in the field of smart IoT parking systems and educate the technological community about the technologies, features, and procedures for implementing the smart parking management system. In the paper, we aim to summarise the works on occupancy monitoring in smart IoT parking systems addressing the advantages and issues of the present occupancy monitoring methods, also suggesting future inclusions for smart IoT parking systems.

Automatic vehicle identification from Google Earth satellite images based on single shot deep learning neural networks for object detection

Automatic vehicle identification from Google Earth satellite images based on single shot deep learning neural networks for object detection

Automated Vehicle Identification Based on Car-Following Data With Machine Learning

Automated Vehicle Identification Based on Car-Following Data With Machine Learning

Licence Plate Detection Using Machine Learning

License Plate Recognition (LPR) is one of the tough tasks in the field of computer vision. Although it has been around for quite a while, there still lies the challenges when we have to deal with; the harsh environmental conditions like snowy, rainfall, windy, low light conditions etc. as well as the condition of the plates which includes the bent, rotated, broken plates. The performance of the recognition and detection frameworks take a significant hit when it is concerned with these conditional effects on the license plate. In this paper, we introduced a model to improve our accuracy based on the Chinese Car Parking Dataset (CCPD) using 2 separate convolutional neural networks. The first CNN will be able to detect the bounding boxes for the license plate detection using Non-Maximus Suppression (NMS) to find the most probable bounding area whereas the second one will take these bounding boxes and use the spatial attenuation network and character recognition model to successfully recognize the license plate. First, we train the CNN to detect the license plates, then use the second CNN to recognize the characters. The overall recognition accuracy was found to be 89% in the CCPD dataset.

License Plate Recognition: A Brief Overview

Abstract: The License Plate Recognition system is a technological solution designed to automatically capture and interpret license plate information from vehicles. Utilizing a blend of object detection, character segmentation, optical character recognition (OCR) and image processing, ANPR systems play a pivotal role in bolstering security, optimizing traffic management, and supporting diverse applications such as toll collection and parking management. Equipped with highresolution cameras and advanced algorithms, these systems are integral to modern surveillance and transportation, providing solutions that enhance safety and operational efficiency.

Using acoustic camera technology on inspection of noisy vehicles in Taiwan

In recent years, noise complaints on motor vehicles had shown an increasing trend and became an environmental concern issue in Taiwan. Since 2021, innovative acoustic camera technology used to inspect high-noise vehicles exceeding "Vehicle Noise Control Standard" would be fined. There are 111 sets of acoustic camera equipment using in 21 counties and cities in Taiwan, and local government issued 4,099 fines to violating vehicle owners. This environmental policy has been supported by people living near major road with ninety percent satisfaction. Currently, acoustic camera technology with sound radar adopted by some cities in France to locate noisy vehicles accurately and issue fines automatically. Array sound camera technology has been used to inspect noisy vehicles in Netherlands and Germany. Taiwan is the first country in Asia to use acoustic camera technology conducting inspections and law enforcement against noisy vehicles. The sound camera technology system is mainly composed of high-speed camera equipment, sound level meter and license plate recognition system. Fixed systems are prior to set on specific road in hot spots of public complaint to inspect and outlaw noisy vehicles. After implementation of acoustic camera technology in these years, some challenges need to overcome. First, people doubt with accuracy and fairness of law enforcement equipment. Second, noise judgment of multi-source vehicle with law enforcement. To improve accuracy of law enforcement and multi-sound source vehicle noise judgment, technology of combining sound photography with array microphones is developing in Taiwan. In the future, we plan to conduct research on multi-source vehicle noise judgment technology to improve inspection efficiency and ensure a peaceful environment.

Clustering pipeline for vehicle behavior in smart villages

Smart cities and villages present a plethora of opportunities for fusing and managing multi-source data. However, in the analysis of mobility patterns, the use of only one data source (i.e., road sensors) without considering other contextual data sources, limits the understanding of the process. To address this gap, we propose a pipeline that integrates multiple data sources, providing valuable information for pattern extraction, mainly based on vehicle mobility behavior and provenance. Our research also highlights the critical role of selecting the appropriate normalization algorithm to scale input features from heterogeneous data sources, which has not received sufficient attention in the literature. We conducted our analysis using data from four License Plate Recognition (LPR) cameras, spanning nine months, and incorporating several databases that include provenance, gross income, and holiday information, resulting in a dataset of over 50,000 vehicles. Using this data and our clustering pipeline, we identified various traffic patterns among residents and visitors in a rural touristic area. Our findings assist data analysts in choosing algorithms for analyzing heterogeneous datasets. Moreover, policymakers could use our results to adjust the resources, such as new parking zones.

Objective Video Quality Assessment and Ground Truth Coordinates for Automatic License Plate Recognition

In the realm of modern video processing systems, traditional metrics such as the Peak Signal-to-Noise Ratio and Structural Similarity are often insufficient for evaluating videos intended for recognition tasks, like object or license plate recognition. Recognizing the need for specialized assessment in this domain, this study introduces a novel approach tailored to Automatic License Plate Recognition (ALPR). We developed a robust evaluation framework using a dataset with ground truth coordinates for ALPR. This dataset includes video frames captured under various conditions, including occlusions, to facilitate comprehensive model training, testing, and validation. Our methodology simulates quality degradation using a digital camera image acquisition model, representing how luminous flux is transformed into digital images. The model’s performance was evaluated using Video Quality Indicators within an OpenALPR library context. Our findings show that the model achieves a high F-measure score of 0.777, reflecting its effectiveness in assessing video quality for recognition tasks. The proposed model presents a promising avenue for accurate video quality assessment in ALPR tasks, outperforming traditional metrics in typical recognition application scenarios. This underscores the potential of the methodology for broader adoption in video quality analysis for recognition purposes.

AFA-Net: Adaptive Feature Attention Network in image deblurring and super-resolution for improving license plate recognition

Although a number of license plate recognition (LPR) systems have become significantly advanced, they are still far from producing ideal images. They are still fiddled with low-resolution (LR) and motion blur, which are two of the most common problems in images extracted from automobile driving environment. In order to address this issue, we present a novel LPR method that processes LR and motion blurred images from dash cams. We propose a unique framework, AFA-Net (Adaptive Feature Attention Network) organized by image pre-restoration, feature composition and image reconstruction modules. The proposed AFA-Net can generate a clear restoration image for robust LPR performance with the images obtained from dash cams in an unconstrained environment. Furthermore, we explore the novel problem, Joint-IRLPRNet (Joint-Image Restoration and License Plate Recognition Network), that simultaneously address image restoration (i.e. SR and deblurring) and LPR in an end-to end trainable manner. Moreover, we introduce a dataset called LBLP (LR and blurred license plate (LP)). The dataset is composed of 2,779 LR and motion blurred cropped LP images, extracted from unconstrained dash cams. The experimental results on LBLP dataset indicate that AFA-Net achieves 15.28% improvement in recognition accuracy, 6.47% in sequence similarity, and 3.89% in character similarity, compared to the traditional LPR model with image restoration model. Moreover, Joint-IRLPRNet can be more effective results than AFA-Net.

A generative adversarial network-based framework for network-wide travel time reliability prediction

This paper introduces a generative model named the travel time reliability-generative adversarial network (TTR-GAN) model for predicting network-wide TTR using automatic vehicle identification data. The TTR-GAN model is capable of generating predicted travel time samples, enabling the assessment of network-wide TTR without the need to assume a specific travel time distribution. In the TTR-GAN model, a combination of graph convolutional networks and long short-term memory (LSTM) neural networks is employed within the GAN framework. When training the TTR-GAN model, special attention is given to adjusting the mean and standard deviation of the generated samples, aiming for a closer resemblance to real samples. Experiments conducted on a road network in China demonstrate the predictive capability of the proposed TTR-GAN model, surpassing several benchmark models such as the LSTM neural network, moving average model, and GAN model in terms of statistical, buffer time, and probability distribution measures. By incorporating the mean and standard deviation into the loss function, the TTR-GAN model achieves an 18.2% reduction in Jensen‒Shannon divergence between predicted and real samples. Furthermore, the model's performance in real-world applications is illustrated through a sensitivity test.

Suppression of laser speckle contrast using negative dielectric anisotropy nematic liquid crystal doped with an ionic medium

ABSTRACT A liquid crystal (LC) with negative dielectric anisotropy called N-(4-Methoxybenzylidene)-4-Butylaniline (MBBA) doped with the ionic agent cetyltrimethylammonium bromide (CTAB) has been investigated as a speckle suppression device to reduce the laser’s speckle contrast. The electrohydrodynamic instabilities (EHDI) effect in LC is exploited to reduce speckle contrast. The EHDI effect causes chaotic turbulence in LC, resulting in dynamic scattering of the incident light beam. In order to investigate the electro-optic properties and frequency region for optimal performance, optical measurements and dielectric spectroscopy of the pristine and doped LC are performed. The dielectric characteristics of the LC demonstrate that the frequency range to produce the EHDI effect in the LC cell is broadened as the ionic agent is doped. The inclusion of ions also enhances the ion density, mobility, and conductivity in LC. The increase in ion density and mobility enhances the turbulence in the LC at a faster rate; as a result, the dynamic scattering becomes more rapid. When measured with a charge-coupled device (CCD) camera, the faster dynamic scattering provides higher suppression in the laser’s speckle contrast. This method can be used to reduce the speckle contrast in laser source-based imaging and projection displays.

Disentangled generation network for enlarged license plate recognition and a unified dataset

License plate recognition plays a critical role in many practical applications, but license plates of large vehicles are difficult to be recognized due to the factors of low resolution, contamination, low illumination, and occlusion, to name a few. To overcome the above challenges, the transportation management department generally introduces the enlarged license plate behind the rear of a vehicle. However, enlarged license plates have high diversity as they are non-standard in position, size, and style. Furthermore, the background regions contain a variety of noisy information which greatly disturbs the recognition of license plate characters. In this work, we address the enlarged license plate recognition problem and contribute a dataset containing 9342 images, which cover most of the challenges of real scenes. However, the created data are still insufficient to train deep methods of enlarged license plate recognition, and building large-scale training data is very time-consuming and high labor cost. To handle this problem, we propose a novel data generation framework based on the Disentangled Generation Network (DGNet), which disentangles the generation of enlarged license plate data into the text generation and background generation in an end-to-end manner to effectively ensure the generation diversity and integrity, for robust enlarged license plate recognition. Extensive experiments on the created dataset are conducted, and we demonstrate the effectiveness of the proposed approach in three representative text recognition frameworks.

A Light Vehicle License-Plate-Recognition System Based on Hybrid Edge–Cloud Computing

With the world moving towards low-carbon and environmentally friendly development, the rapid growth of new-energy vehicles is evident. The utilization of deep-learning-based license-plate-recognition (LPR) algorithms has become widespread. However, existing LPR systems have difficulty achieving timely, effective, and energy-saving recognition due to their inherent limitations such as high latency and energy consumption. An innovative Edge–LPR system that leverages edge computing and lightweight network models is proposed in this paper. With the help of this technology, the excessive reliance on the computational capacity and the uneven implementation of resources of cloud computing can be successfully mitigated. The system is specifically a simple LPR. Channel pruning was used to reconstruct the backbone layer, reduce the network model parameters, and effectively reduce the GPU resource consumption. By utilizing the computing resources of the Intel second-generation computing stick, the network models were deployed on edge gateways to detect license plates directly. The reliability and effectiveness of the Edge–LPR system were validated through the experimental analysis of the CCPD standard dataset and real-time monitoring dataset from charging stations. The experimental results from the CCPD common dataset demonstrated that the network’s total number of parameters was only 0.606 MB, with an impressive accuracy rate of 97%.

Detection of Vehicle Number Plate and Speed Using Machine Learning

In an era marked by a significant upsurge in vehicular traffic and the resultant increase in road accidents, the need for advanced traffic management systems is paramount. This paper presents an innovative solution, "Detection of vehicle number plate and speed using machine learning" which aims to enhance road safety by addressing the persistent problem of over-speeding. The project leverages state-of-the-art technologies, including machine learning, deep learning, and computer vision, to develop a smart system that can detect speeding vehicles, recognize license plates, and record crucial data related to speed limit violations. The primary motivation for this endeavor is the escalating rate of road accidents, particularly in India, driven by speeding in areas susceptible to accidents. The system employs image segmentation, corner detection algorithms, filtering algorithms, and automatic vehicle plate recognition techniques to achieve its objectives. By using machine learning and deep learning models, the system accurately identifies license plates, segments characters on these plates, and recognizes them. In summary, this project represents a significant advancement in road safety and traffic management. By deploying an intelligent system that can proactively detect over-speeding vehicles and maintain comprehensive records, it strives to create a safer and more responsible driving environment, ultimately working towards the prevention of accidents and saving lives. Key Words: Automatic Number Plate Recognition, Vehicle Speed Detection, Machine Learning, Computer Vision, Traffic Management.