Academic literature on the topic 'LICENSE PLATE DETECTION SYSTEM'

This thesis evaluates the performance of a vehicle detection technology, Automated License Plate Recognition (ALPR) camera systems, with regards to its ability to produce real-time travel time information in active work zones. A literature review was conducted to investigate the ALPR technology as well as to identify other research that has been conducted using ALPR systems to collect travel time information. Next, the ALPR technology was tested in a series of field deployments in both an arterial and a freeway environment. The goal of the arterial field deployment was to evaluate the optimal ALPR camera angles that produce the highest license plate detection rates and accuracy percentages. Next, a series of freeway deployments were conducted on corridors of I-285 in Atlanta, Georgia in order to evaluate the ALPR system in active work zone environments. During the series of I-285 freeway deployments, ALPR data was collected in conjunction with data from Bluetooth and radar technologies, as well as from high definition video cameras. The data collected during the I-285 deployments was analyzed to determine the ALPR vehicle detection rates. Additionally, a script was written to match the ALPR reads across two data collection stations to determine the ALPR travel times through the corridors. The ALPR travel time data was compared with the travel time data produced by the Bluetooth and video cameras with a particular focus on identifying travel time biases associated with each given technology. Finally, based on the knowledge gained, recommendations for larger-scale ALPR work zone deployments as well as suggestions for future research are provided.

<p> In this work, we develop a license plate detection method using a SVM (Support Vector Machine) classifier with HOG (Histogram of Oriented Gradients) features. The system performs window searching at different scales and analyzes the HOG feature using a SVM and locates their bounding boxes using a Mean Shift method. Edge information is used to accelerate the time consuming scanning process. </p><p> Our license plate detection results show that this method is relatively insensitive to variations in illumination, license plate patterns, camera perspective and background variations. We tested our method on 200 real life images, captured on Chinese highways under different weather conditions and lighting conditions. And we achieved a detection rate of 100%. </p><p> After detecting license plates, alignment is then performed on the plate candidates. Conceptually, this alignment method searches neighbors of the bounding box detected, and finds the optimum edge position where the outside regions are very different from the inside regions of the license plate, from color's perspective in RGB space. This method accurately aligns the bounding box to the edges of the plate so that the subsequent license plate segmentation and recognition can be performed accurately and reliably. </p><p> The system performs license plate segmentation using global alignment on the binary license plate. A global model depending on the layout of license plates is proposed to segment the plates. This model searches for the optimum position where the characters are all segmented but not chopped into pieces. At last, the characters are recognized by another SVM classifier, with a feature size of 576, including raw features, vertical and horizontal scanning features. </p><p> Our character recognition results show that 99% of the digits are successfully recognized, while the letters achieve an recognition rate of 95%. </p><p> The license plate recognition system was then incorporated into an embedded system for parallel computing. Several TS7250 and an auxiliary board are used to simulate the process of vehicle retrieval.</p>

CNPq<br>Sistemas de controle de velocidade são utilizados em vários países para fiscalizar o cumprimento dos limites de velocidade, prevenindo assim acidentes de trânsito. Muitos desses sistemas são baseados em tecnologias intrusivas que requerem processos de instalação e manutenção complexos, geralmente atrapalhando o trânsito. Neste projeto, propõe-se um sistema não intrusivo para estimativa da velocidade de veículos baseado em vídeo. O sistema proposto detecta veículos em movimento utilizando um detector de movimento otimizado. Aplicou-se um detector de texto especializado para localizar a placa dos veículos, a qual foi utilizada para seleção e rastreamento de pontos estáveis. Os pontos rastreados são então filtrados e retificados para remoção do efeito da perspectiva. A velocidade dos veículos é estimada comparando-se a trajetória dos pontos rastreados com dimensões conhecidas no mundo. Para os testes, utilizou-se aproximadamente cinco horas de vídeos em diferentes condições, capturados por uma câmera de baixo custo posicionada a 5,5 metros de altura. Os vídeos capturados contém mais de 8.000 veículos distribuídos em três pistas diferentes, com as velocidades reais para cada veículo obtidas a partir de um detector por laço indutivo. O detector de placas proposto foi comparado com três outros métodos no estado da arte e obteve os melhores resultados de performance para os nossos vídeos, com precisão (precision) de 0,93 e coeficiente de revocação (recall) de 0,87. A estimativa de velocidade dos veículos apresentou erro médio de -0,5 km/h, permanecendo dentro da margem de +2/-3 km/h, determinada por agências reguladoras em vários países, em 96,0% dos casos.<br>Speed control systems are used in most countries to enforce speed limits and, consequently, to prevent accidents. Most of such systems are based on intrusive technologies which require complex installation and maintenance, usually causing traffic disturbance. In this work, we propose a non-intrusive video-based system for vehicle speed estimation. The proposed system detects moving vehicles using an optimized motion detector. We apply a specialized text detector to locate the vehicle’s license plate region, in which stable features are selected for tracking. The tracked features are then filtered and rectified for perspective distortion. Vehicle speed is estimated by comparing the trajectory of the tracked features to known real world measures. For our tests, we used almost five hours of videos in different conditions, captured by a single low-cost camera positioned at 5.5 meters height. The recorded videos contain more than 8,000 vehicles, in three different road lanes, with associated ground truth speeds obtained from an inductive loop detector. We compared our license plate detector with three other state-of-the-art text detectors, and our approach has shown the best performance for our dataset, attaining a precision of 0.93 and a recall of 0.87. Vehicle speeds were estimated with an average error of -0.5 km/h, staying inside the +2/-3 km/h limit determined by regulatory authorities in several countries in over 96.0% of the cases.

License Plate Recognition (LPR) systems are frequently utilized in various access controls and security applications. In this thesis, an experimental constraint based real-time License Plate Recognition system is designed, and implemented in Java platform. Many of the available constraint based methods worked under strict restrictions such as plate color, fixed illumination and designated routes, whereas, only the license plate geometry and format constraints are used in this developed system. These constraints are built on top of the current Turkish license plate regulations. The plate localization algorithm is based on vertical edge features where constraints are used to filter out non-text regions. Vertical and horizontal projections are used for character segmentation and Multi Layered Perceptron (MLP) based Optical Character Recognition (OCR) module has been implemented for character identification. The extracted license plate characters are validated against possible license plate formats during the recognition process. The system is tested both with Turkish and foreign license plate images including various plate orientation, image quality and size. An accuracy of 92% is achieved for license plate localization and %88 for character segmentation and recognition.

碩士<br>玄奘大學<br>資訊管理學系碩士班<br>98<br>Autonomous surveillance systems are widely used as an important tool for security control in public areas. Among the numerous targets of the autonomous surveillance system, license plate recognition can help to identify the terrorist in cars. Uneven lighting conditions happen all the time in an autonomous surveillance system. As a result, traditional license plate detection can’t achieve the goal. In this research, we propose a novel method for license plate localization. The features of a license plate is generated in preprocess through the morphology. Then, apply the AdaBoost algorithm to select some weak classifiers from the weak classifier space to construct a strong classifier. Experimental results show that the proposed method can efficiently detect license plates under different illuminations.

碩士<br>國立雲林科技大學<br>電機工程系碩士班<br>92<br>The amelioration of social order in our country caused by gun-fire event, extortion of car and human hostage, and robberies has greatly influence the society. This crisis has made security check and video surveillance become more important. Currently, the installation of video surveillance cameras is getting popular on the streets of many counties and cities. The recording of the passing vehicles and their license plate in these videos have successfully help the police solving many cases including hit-and-run, extortion of human hostage, etc.. However, search the video to extract and record the license plate (LP) of the suspected vehicles needs a lot of effort and time. This tedious work sometimes decreases the willingness and efficiency of the police in pursuing the criminals which places a great threat to our society. It''s obvious that the police needs an automatic system to extract the license plate from the video for them. In view of the above demand, we propose utilizing the techniques of image processing and develop an automatic LP detection system to suit the police department''s urgent need. Although most of current video cameras are equipped with a resolution of 640x480, to save the storage space, videos are usually taken at low resolution (320x240) and saved at high compression ratio. This available low resolution and poor quality data not only distinguish the research in this thesis from that of the other LP detection system, it also puts a great challenge to our design. In this thesis, we deal with both daytime and night-time video. On the basis of local variance of the image and techniques of morphology (e.g. bottom-hat, top-hat), method which detect location of the license plate was successfully developed. Videos of 12-hour daytime and 2-hour night-time taken by police department from several street scenes are used to evaluate the system performance. The results show that the LP detection rates are 93.74% and 78.57% for the daytime and night-time, respectively. Their corresponding false alarm rates are 1.17% and 15.58%.

碩士<br>國立臺南大學<br>數位學習科技學系碩士班<br>96<br>The paper proposed an autonomous license plate detection and recognition system with computer vision. The system consists of four subsystems: car detection subsystem, plate extraction subsystem, character division subsystem and character recognition subsystem. Car detection subsystem uses MMADR and NDDR of dynamic image to find the location of the cars on the screen. Plate extraction subsystem uses the characteristics of the plates and algorithm used to search plates to extract plate; character division subsystem combines Tophat, Labeling and LRE to automatically divide. As to character recognition subsystem, after comparing identification effects of SVM and BPNN, we choose BPNN as the recognizer. Experiment outcome proves that our system can effectively detect cars and recognize the plates under different lights.

碩士<br>中華技術學院<br>電子工程研究所碩士班<br>98<br>The purpose of this research was to implement License Plate Detection System by C Language, in which MathWork MATLAB 7.6 and Microsoft Visual Studio 2005 were main development tools. Mex-function was implemented in C Language since it is essentially famous for efficiency, structure, portability and good readability. This study tries to take advantage of MATLAB API to dynamically link MEX library in run-time. The goal is turning script architecture (M-files) into Mex-functions. Programming process adopted a progressive approach to retain the benefits of MATLAB workspace, which means MATLAB built-in instructions are gradually replaced step by step. The method is able to easily verification and evaluation the performance by MATLAB’s friendly user interface. Photos taken from the digital camera provide the input images to the system in which environmental conditions could be day or night without external light source. Default accepted resolution is 1024x768 pixels. Experimental results showed that as long as the pixel numbers of license plate close to assumed 60x220, the characters could be recognized successfully. With appropriate tunned parameters, it can support more image sizes and variety of different plate pixels. The final system output is an image that re-combinated by the prefabricated digit/alphabet pictures according to the characters identification result from original images passed through processes of the license plate location, template matching and some pre-treatment.

碩士<br>國立中興大學<br>電機工程學系所<br>102<br>License Plate Recognition (LPR) has recently played a critical role in applications of Intelligent Transportation System (ITS), such as parking spaces management, records of traffic violations, and searches for stolen cars. However, although research on license plate recognition has been undertaken over many years, in practical applications, the accuracy of detection and identification can decrease because of the complex backgrounds and light changes in capture image. Therefore, recognition systems should be limited to specific environmental settings to obtain favorable results. In summary, interference caused by the environment is the main factor that limits the development of LPR. LPR and License Plate Detection (LPD) comprise two main steps, character segmentation and character identification. The license plate zone is detected prior to the two steps, and is the key to successful recognition of the license plate zone. Most LPR studies have investigated static and low-resolution images. Conversely, the present study involved using 720P high-definition video and a PCIe (PCI Express) image capture card to detect license plates. Subsequently, to achieve real-time LPR, Haar Wavelet Transform was applied in extracting an HL image, reducing the algorithm calculation time. Chapter 1 provides a review of relevant literature, and introduces the system architecture developed in this study. Chapter 2 introduces related techniques, including color space transform, motion detection, image sharpening, histogram equalization, Haar wavelet transform, edge detection, binarization, and connected component labeling. Chapter 3 describes the algorithm used for LPD in detail. First, the motion object is detected. The license plate features are then intensified, and adaptive thresholding is conducted to binarize the image. The edge is then connected, and connected component labeling is used to label each block. Thus, the license plate location is obtained. Chapter 4 introduces the experimental equipment and software interface used. In addition the experimental results obtained using the aforementioned algorithm in this study are described. In the final chapter, a conclusion is presented and possible methods of the amendments are suggested in accordance with the experimental results.

A continual upsurge in the volume of vehicles has been noticed over the past few decades with the increase in population all over the world. Therefore, tracking of vehicles depending upon the number plates is crucial to guarantee the control of vehicular traffic in competent manner. The vehicles can be detected on the basis of their tags with the help of a new image processing-based technology referred as ANPR (Automatic Number Plate Recognition) the expertise is ahead of time ubiquity to ensure security and traffic management. This system makes use of computer vision approach for extracting information regarding the abnormal state from a digital image using a computer. Almost all number plate localization algorithms combine many processes that result in a long computational time. Most of the image details are lost or image quality gets degraded as a result of complex, noisy content in images. The non-consistency of processes cause degradation which in turn affects the image quality. The car number plate detection has many stages. In this research work, technique of voting classifier is used for detecting the number plates of cars. For the purpose of voting classification, we have used a unique combination of classifiers. The voting classification proposed in this research work for the number plate detection is the combination of SVM and random forest classifier. The MATLAB and/or GNU Octave has been used for the evaluation of the proposed model. The efficiency of new algorithmic approach is examined with respect to accuracy, precision and recall. The proposed algorithm gives accuracy up to 95 percent for the car number plate detection. Similar, observation with the Precision and the Recall that comes out to be 95.81 percent and 95.45 percent respectively.