Dissertations / Theses on the topic 'Car license plate detection and recognition'
To see the other types of publications on this topic, follow the link: Car license plate detection and recognition.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 37 dissertations / theses for your research on the topic 'Car license plate detection and recognition.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
Ning, Guanghan. "Vehicle license plate detection and recognition." Thesis, University of Missouri - Columbia, 2016. http://pqdtopen.proquest.com/#viewpdf?dispub=10157318.
Full textAbstract:
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.
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%.
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.
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.
Our character recognition results show that 99% of the digits are successfully recognized, while the letters achieve an recognition rate of 95%.
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.
2
Colberg, Kathryn. "Investigating the ability of automated license plate recognition camera systems to measure travel times in work zones." Thesis, Georgia Institute of Technology, 2013. http://hdl.handle.net/1853/49048.
Full textAbstract:
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.
3
Krajíček, Pavel. "Rozpoznání SPZ/RZ." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2010. http://www.nusl.cz/ntk/nusl-218307.
Full textAbstract:
The theme of this thesi’s deals with the detection and recognition of car license plate from pictures made of screening machine situated on a crassing or inside a car. The thesis si divided into two basic parts. First deals with searching for presence of licence plate in the picture. If the marque was found, we continue the second part of the program which identificates the found license plate. The first part of program aspires to find the licence plate by the edge detectors. The second part classifies characters by the method based on an analytical description.
4
D'amore, Luiz Angelo. "Robustez na segmentação de placas veiculares em condições complexas de aquisição." Universidade Presbiteriana Mackenzie, 2010. http://tede.mackenzie.br/jspui/handle/tede/1389.
Full textAbstract:
Made available in DSpace on 2016-03-15T19:37:29Z (GMT). No. of bitstreams: 1
Luiz Angelo D Amore.pdf: 3689058 bytes, checksum: 8476274d8f5220a2a7978da28a4a4f3d (MD5)
Previous issue date: 2010-08-13Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
The work presented here shows a robust method for license plate detection. The term robust in this work is directly related to the efficacy of the system as an automated locator of license plates without human intervention and considering specific characteristics of image acquisition and license plate features. The proposed method is based on the specify features of the digits found on the Brazilian license plates. Although the method was designed for the Brazilian license plate pattern it can be easily adjusted to other patterns. The results obtained using the proposed method showed a better performance than that of other academic approaches and even of commercial systems.
Os sistemas automáticos de reconhecimento de placas veiculares têm como principal função a identificação de veículos a partir de imagens digitais, com aplicações nas áreas de segurança pública e privada. Neste trabalho são apresentadas técnicas de processamento de imagens com o objetivo de desenvolver um método robusto para a segmentação de placas veiculares em condições complexas de aquisição. O termo robusto neste trabalho é relacionado diretamente à eficácia do sistema quanto à localização automática das placas veiculares sem intervenção humana, considerando características específicas das imagens e placas. O método proposto é baseado nas especificidades dos dígitos localizados nas placas brasileiras. Embora o método tenha sido projetado para o padrão de placas brasileiro, pode ser facilmente ajustado para outros padrões. Os resultados obtidos com o método proposto mostram um desempenho melhor que outras abordagens acadêmicas, ou mesmo de sistemas comerciais.
5
Vladimir, Tadić. "Fazifikacija Gaborovog filtra i njena primena u detekciji registarskih tablica." Phd thesis, Univerzitet u Novom Sadu, Fakultet tehničkih nauka u Novom Sadu, 2018. https://www.cris.uns.ac.rs/record.jsf?recordId=107171&source=NDLTD&language=en.
Full textAbstract:
Disertacija prikazuje novi algoritam za detekciju i izdvajanje registarskih tablica iz slike vozila koristeći fazi 2D Gaborov filtar. Parametri filtra: orijentacija i talasna dužina su fazifikovani u cilju optimizacije odziva Gaborovog filtra i postizanja dodatne selektivnosti filtra. Prethodno navedeni parametri dominiraju u rezultatu filtriranja. Bellova i trougaona funkcija pripadnosti pokazale su se kao najbolji izbor pri fazifikaciji parametara filtra. Algoritam je evaluiran nad više baza slika i postignuti su zadovoljavajući rezultati. Komponente od interesa su efikasno izdvojene i postignuta značajna otpornost na šum i degradaciju na slici.The thesis presents a new algorithm for detection and extraction of license plates from a vehicle image using a fuzzy two-dimensional Gabor filter. The filter parameters, orientation and wavelengths are fuzzified to optimize the Gabor filter’s response and achieve a greater selectivity. It was concluded that Bell’s function and triangular membership function are the most efficient methods for fuzzification. Algorithm was evaluated on several databases and has provided satisfactory results. The components of interest were efficiently extracted, and the procedure was found to be very noise-resistant.
6
Špaňhel, Jakub. "Re-identifikace vozidla pomocí rozpoznání jeho registrační značky." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2015. http://www.nusl.cz/ntk/nusl-264932.
Full textAbstract:
This thesis aims at proposing vehicle license plate detection and recognition algorithms, suitable for vehicle re-identification. Simple urban traffic analysis system is also proposed. Multiple stages of this system was developed and tested. Specifically - vehicle detection, license plate detection and recognition. Vehicle detection is based on background substraction method, which results in an average hit rate of ~92%. License plate detection is done by cascade classifiers and achieves an average hit rate of 81.92% and precision rate of 94.42%. License plate recognition based on Template matching results in an average precission rate of 60.55%. Therefore the new license plate recognition method based on license plate scanning using the sliding window principle and neural network recognition was introduced. Neural network achieves a precision rate of 64.47% for five input features. Low precision rate of neural network is caused by small amount of training sample for some specific license plate characters.
7
Li, Hui. "Text detection and recognition in natural scene images." Thesis, 2018. http://hdl.handle.net/2440/115175.
Full textAbstract:
This thesis addresses the problem of end-to-end text detection and recognition in
natural scene images based on deep neural networks. Scene text detection and recognition
aim to find regions in an image that are considered as text by human beings,
generate a bounding box for each word and output a corresponding sequence of
characters. As a useful task in image analysis, scene text detection and recognition
attract much attention in computer vision field. In this thesis, we tackle this problem
by taking advantage of the success in deep learning techniques.
Car license plates can be viewed as a spacial case of scene text, as they both consist
of characters and appear in natural scenes. Nevertheless, they have their respective
specificities. During the research progress, we start from car license plate detection
and recognition. Then we extend the methods to general scene text, with additional
ideas proposed.
For both tasks, we develop two approaches respectively: a stepwise one and
an integrated one. Stepwise methods tackle text detection and recognition step by
step by respective models; while integrated methods handle both text detection and
recognition simultaneously via one model. All approaches are based on the powerful
deep Convolutional Neural Networks (CNNs) and Recurrent Neural Networks
(RNNs), considering the tremendous breakthroughs they brought into the computer
vision community.
To begin with, a stepwise framework is proposed to tackle text detection and
recognition, with its application to car license plates and general scene text respectively.
A character CNN classifier is well trained to detect characters from an image
in a sliding window manner. The detected characters are then grouped together as
license plates or text lines according to some heuristic rules. A sequence labeling
based method is proposed to recognize the whole license plate or text line without
character level segmentation.
On the basis of the sequence labeling based recognition method, to accelerate the
processing speed, an integrated deep neural network is then proposed to address
car license plate detection and recognition concurrently. It integrates both CNNs
and RNNs in one network, and can be trained end-to-end. Both car license plate
bounding boxes and their labels are generated in a single forward evaluation of the
network. The whole process involves no heuristic rule, and avoids intermediate
procedures like image cropping or feature recalculation, which not only prevents
error accumulation, but also reduces computation burden.
Lastly, the unified network is extended to simultaneous general text detection and
recognition in natural scene. In contrast to the one for car license plates, some innovations
are proposed to accommodate the special characteristics of general text. A
varying-size RoI encoding method is proposed to handle the various aspect ratios of general text. An attention-based sequence-to-sequence learning structure is adopted
for word recognition. It is expected that a character-level language model can be
learnt in this manner. The whole framework can be trained end-to-end, requiring
only images, the ground-truth bounding boxes and text labels. Through end-to-end
training, the learned features can be more discriminative, which improves the overall
performance. The convolutional features are calculated only once and shared by both
detection and recognition, which saves the processing time. The proposed method
has achieved state-of-the-art performance on several standard benchmark datasets.Thesis (Ph.D.) -- University of Adelaide, School of Computer Science, 2018
8
tsai, Sung-nien, and 蔡松年. "Dynamic Car License Plate Detection." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/22083988530781627442.
Full textAbstract:
碩士中華技術學院
電子工程研究所碩士班
98
License plate recognition system is widely used in a lot of areas such as in the automation of parking lot toll station and in helping to detect stolen vehicle. Previous studies are essentially in static system needed to use a image in which the license plate is fangzheng, however, in this way the vehicle's location and the mobility and the environment must be satisfied some conditions. This study focuses on the dynamic vehicle license plate recognition using video car image when car is moving. In the first, the source images are filmed from the camera. Then, make a series of processes to the images as filtering, edge detection, binarization, rotation…etc. And then, using the binary images locates the preliminary position of the license plate. Finally, a template matching method is to be used to obtain more precise position of the license plate location. The actual recording films of the moving vehicle are used to test. The results of the test verify the effectiveness of the proposed method.
9
Chen, I.-Chih, and 陳奕志. "Constructing Embedded Car License Plate Recognition System." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/91717196909075534200.
Full textAbstract:
碩士淡江大學
資訊工程學系碩士班
93
Embedded System is a computing platform designed for specific purpose. Because it’s task is much simpler than personal computer which is designed for general computing purpose, Embedded System can simplify its hardware architecture, cost down its hardware price, produce smaller device and low energy consumption. It also fits to be mobile computing platform. But Embedded System is constrained by its simple architecture; its processing power is much slower than personal computer. This paper purpose a software porting procedure between personal computer and Embedded System platform via a instance of Car License Plate Recognition System , and make Embedded Car License Plate Recognition System more efficient via exchange floating operation by integer operation and bitwise operation. For Example the image format captured by COMS, is Color Filter Array ,and this format will lost 2/3’s original image illumination. The illumination recovery process was originally involve with mass of floating operation, after applying the speeding method that we just mention before the processing time become times faster. We also use uClinux to assist hardware communication and process scheduling, and uClinux makes Embedded System able to handle complicated process control, also make software porting much smooth. Network File System ,NFS not only resolving the problem of lacking storage media, but also reduces the times of flash Rom burning procedure. Finally we use TCP/IP to transfer the image captured by CMOS to remote personal computer for Car License Plate Recognition ,and compare Car License Plate Recognition results between Embedded System and personal computer.
10
Wang, Ching-Chung, and 王精忠. "THE STUDY OF CAR LICENSE PLATE RECOGNITION SYSTEM." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/84668349117339820560.
Full textAbstract:
碩士大同大學
通訊工程研究所
93
Along with economical grow up and commerce activity vigorous development, people for the automobile need is more and more, although government for the traffic construction is very popular, but in the crowded Taiwan area, the question of parking space not enough is a fact of without saying, so how to manage parking lots efficiently and increasing usability of the parking lots that is our concerned question. This thesis proposed the license plate recognition system, includes license plate locating, image binarization, calibration of license plate, character segmentation, character recognition and so on, total five parts; In the license plate locating, we use the image process technique to process the input image of automobile change into fixed resolution gray image, use again Sobel edge detection method to find out the edge of license plate, at last use filter to find out the position of license plate; In the image binarization, we use dynamic threshold value method to find out threshold value, let gray image of license plate change into binarized image; In the calibration of license plate, we use bottom outline of license plate analysis method to find out slope angle of license plate and to execute calibration; In the character segmentation, we use vertical projection method to find out the high of character, and we use horizontal projection method to segment the characters of license plate, at last we use partial recognition method to recognize the number of license plate image. This system takes 200 license plate images from indoor and outdoor parking lots to execute the experiment of license plate recognition, experimental results, the license plate locating successful rate is 98%, the character segmentation successful rate is 95%, the character recognition successful rate is 93%, the average recognition time of each image needs 1.2 second.
11
Wu, Yi-Hsuan, and 吳宜軒. "A Vision-based Car License Plate Recognition System." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/93542526634672435652.
Full textAbstract:
碩士亞洲大學
資訊工程學系碩士班
97
This paper presents a license plate recognition system based on the computer vision, including five parts which are image pre-processing, license plate locating, character segmentation and recognition. In the image pre-processing part, the RGB images obtained from the camera are changed into the gray images. After the gray images are changed into the binaries images, and the noises are filtered by using the median filter. In the license plate locating part, we predict the possible location of the license plate in the binaries image and acquire it with using the length and width value of the general license plate. In the character segmentation part, the characters of the license plate are segmented due to the high contrast between those and the background. At last the neural network is adopted into the pattern recognition, and the letters and numbers could be recognized. Moreover, we exploit the Bottom-hat module to enhance the features of the license plate in order to promote the identification significantly.
12
Ou, Shih-Tsung, and 歐世聰. "The study of car license plate recognition system." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/09289236464793638879.
Full textAbstract:
碩士明新科技大學
電子工程研究所
98
The recognition of license plate system has been resear ched for more than ten years. The recognition rate could not be raised efficiently and be utilized in daily as a result of hardware and software development was unstable in the initial stage. More amd more recognition of license plate system be not applied in daily until hardware and software were developing quickly recently. The most important things to recognition of license plate system are high recognition rate efficiently and take a few time as less as possible. Now all related research are depended on this direction. The paper is that used an improved iterative method binarization as a pre-processing and a continuous filter method be used to clean noise. Then character recognition would be used the way of identification samples that is fast to recognizw. It can be raised its recognition rate, base on a lot of samples. Following the different recognition theorem, however, it’s able to achieve high recognition rate without a large number of samples. For the experimental result, the paper had been taken more than 200 pcs photos of the actual license plates as test sample. The recognition rate is 97.60% and the average it cost to recognize a license plate is only 0.1 second.
13
Chuang, Chia-Lung, and 莊佳龍. "Vehicle Detection and License Plate Recognition." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/57361447713712641354.
Full textAbstract:
碩士國立中正大學
光機電整合工程研究所
93
License plate recognition system has been extensively used in a variety of applications, such as parking lot management of community and buildings, and searching stolen car of police departments, roadway monitor, car management and so on。In the past study, the needed car is catched by the method of responding and activating. In this system, the video camera continuously shoot the motion-based vehicle and the video frames are analyzed by shadow detection, vehicle location, size of license plate to catch the needed image including vehicle automatically. Moreover, a car can be recognized more times according to the characteristics of Multiple-frames to increase the probability of recognition, since the system can be used in reality more practically than which using shooting statically.
14
Chan, Shan-Lun, and 詹尚倫. "Double-line License Plate Detection and Character Segmentation in License Plate Recognition." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/16458793482041748762.
Full textAbstract:
碩士國立交通大學
資訊科學與工程研究所
104
As automatic license plate (LP) localization and recognition getting popular, the requirement of accuracy is rising, too. Most of LP localization systems can be divided into two categories, one is based on technique of image processing and computer vision using edge detection methods such as Sobel operator for gradient computation and search for areas with high texture density as reasonable LP positions; the other is based on machine learning using adaptive boosting with Haar-like features. The accuracy of both methods is close to 100%. Based on the results of localization, character segmentation can be performed by projection and final license plate recognition (LPR) rate up to 95% can be achieved. However, almost all LPR methods are aimed to handle single-line LPs, and may lead to poor recognition rate for the case of double-line LP. To handle such a problem, we propose an image processing method to handle skewed and double-line LP. The proposed approach first performs skew correction of LP and determines whether it has double lines. After line separation, followed by character segmentation performed for each line, single-line LPR can be applied. Experimental results show that the proposed approach is indeed effective in dealing with the situation of skewed LP which may have single or double lines.
15
Shen, Shu-Chun, and 沈淑君. "The detection of car license plate on the road." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/76558277568740855648.
Full textAbstract:
碩士國防大學中正理工學院
電子工程研究所
96
This thesis proposes a car plate detection system to detect the characters from not only a static analysis but also a dynamic analysis for various car plates. The system proposes a real-time algorithm to detect and locate various car plates from roadway images. The system is mounted over the cross-over bridge to capture a sequence of images (image size : 640x480 full color). There are two methods introduced in this study. One is moving object segmentation method and the other is automatic character detection method for car plates. The moving object segmentation adopts the frame difference method and background subtracting method to detect the objects. This is to increase the adaptability for the environment. After turning on the system, the sequent frames can be used to detect the moving objects. The automatic characters detection method of car plates is referred to Multi-Layer Segmentation Method (MLSM). The MLSM method can segment the image depending on the differences of gray level which solves the processing problem of binarization and also deletes noises. Besides, the proposed system has the advantage to extend the applications for checking numbers on the containers of the customhouse because geometry features are used to perform the character detection. And, the detection algorithm will not be influenced by the horizontal or vertical arrangement of characters. Moreover, the proposed algorithm employing the geometry and color relationships of the plates and characters can solve the problems as dealing with the different colors of car plates which have different amounts of characters and will have the character missing problem. This thesis shows good results to detect the characters of the car plates. In the future, the recognition algorithm will be added to recognize the numbers of car plates for the applications of intelligent transportation system.
16
Huang, Hsiu-Ching, and 黃琇靖. "Car License Plate Recognition System in Road Video Camera Application." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/54692254109629351740.
Full textAbstract:
碩士朝陽科技大學
資訊工程系碩士班
95
The intelligent security surveillance system is actively pushed by the governments in Taiwan and other countries in the last few years. With the high technology, it provides the traditional security surveillance system with higher application. Thus a car license plate recognition system based on the digital image processing technique recently becomes a popular research topic. The car license plate recognition system includes three parts: License-Plate Location, Character Extraction, and License-Plate Recognition. In this paper, we derive a new method of the automatic car license plate recognition system in a safely monitored environment. To License-Plate Location, we analyze and describe the features of a car license plate through independent component analysis. We can obtain the features of the car license plate through the independent component analysis filters. The result shows that the average accuracy can be up to 94.3% in different weather and situations. It is higher than the traditional edge-based method which is about 80.6%. We proposed the new dynamic threshold methods for character extraction. It can mostly finish the character extraction in the conditions of shadow variety and contamination of the license plate character. To License-Plate Recognition, we proposed a method to rapidly classify the character based on the structure of character. We can real-time and accurately achieve the character recognition with template matching method. It shows that in the road of high variety environment, the average accuracy of the License-Plate Recognition is 86.3%.
17
Chiang, Chang-Yun, and 江長運. "Plate- and Character-based AdaBoost License Plate Detection and Its Android License Plate Recognition Recorder Implementation." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/86923239518107813208.
Full textAbstract:
碩士國立雲林科技大學
電機工程系碩士班
100
License Plate Detection (LPD) is the fundamental step to License Plate Recognition (LPR). If the license plate localization result of LPD is not exact enough, subsequent OCR step is unnecessary to be carried out. Most especially, on mobile or portable LPR devices, the accuracy of LPD severely suffers from the environmental interferences, like complex background, illumination variation, scene alternation. So most traditional LPR system only can work out for invariant and specific scenes. In order to overcome the LPD difficulty of mobile or portable LPR devices, this paper proposes plate- and character-based AdaBoost LPD with vertical-edge-based auxiliary LPD. The proposed plate-based AdaBoost LPD is to localize 0 or more license plate candidates firstly, and then, the proposed character-based AdaBoost LPD is to verify and filter out the most accurate license plate candidate. As for the vertical-edge-based auxiliary LPD, it is activated to double check the scene only after no license plate candidate is detected by the proposed plate- and character-based AdaBoost LPD. From experimental results, it is found that the proposed LPD method can achieve excellent and stable accuracy under various conditions of background complexity, illumination, and scene. Finally, this paper implements the proposed LPD method and license plate character recognition module into a vehicle video recorder featuring real-time and continuous LPR. The LPD rate and LPR rate of the implementation are 99% and 98%, respectively.
18
Kao, Sho-tsung, and 高碩聰. "An autonomous license plate detection and recognition system." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/72090163125968579734.
Full textAbstract:
碩士國立臺南大學
數位學習科技學系碩士班
96
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.
19
Chen, Ko-Chih, and 陳克智. "License Plate Detection and Recognition of Smart-Phone." Thesis, 2011. http://ndltd.ncl.edu.tw/handle/01428773053617602065.
Full textAbstract:
碩士國立中央大學
資訊工程學系碩士在職專班
99
This paper presents an approach for license plate recognition using a camera-equipped smartphone. The proposed method provides a reliable and accurate technique to solve the problem of license plate recognition caused by the skew and shadow on the license plates. There are four stages in the proposed approach: license plate location, license plate rectification, character segmentation and character recognition. In the first stage, we locate the license plate by accumulating edge points, and then analyze the edge points and accumulation associated with vertical and horizontal dimensions of the image. As to the second stage, license plate rectification, we adopt local threshold to cope with the problem of shadow on the plates first. Next step involved analyzing black and white pixels in order to decide whether to invert the image or not. The researcher tries to engage the characteristics like length-width ratio, size, and position of the bounding box in the text region to eliminate the non-text portions. To solve the rotation, skew, and scale problems of the slanted license plates in the image, we use an affine transformation to estimate the skew angle. Edge points vertical direction accumulating and trough are used to segment characters section in the third stage. We normalize the characters size to 40 × 90. Finally, criterion of normalized cross-correlation is used in the last stage for character recognition. In behalf of shortening the process time for identification, the procedure of character reorganization is improved. We shrink the samples to one-fourth the size to conduct the first identification process. Then, three highest-coefficient samples are chosen to match the original input pattern. From these three samples, the highest-coefficient one is selected as the final result.
20
Lin, Kuo-Chung, and 林國忠. "Implementation for the Recognition of Numbers and Letters of Car License Plate." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/75195583727999834836.
Full textAbstract:
碩士中華技術學院
電子工程研究所碩士班
98
The thesis is focused on the recognition of numbers and letters of car license plate; first of all, the position of the license plate in a photo image must be detected. The thesis proposes that once the position of the license plate is preliminarily detected, a secondary detection method for finding the accurate position of the license plate will be proceeding. The method is to use a template matching method to looking over the entire image to find the maximum values to achieve the accurate position for the facilitation of the subsequent recognition. The templates used in the template matching are built with the sample template database including the numbers from 0 to 9 and the letters from A to Z. Next, using the accurate position obtained from the secondary position detection to segment license plate into a single character, loading and scaling the templates of alphabets and numbers to match the size in accordance with the segmentation of a single character, continually, the sum of difference error square between each template frame and the segmentation of a single character is proceeded one by one to search for the minimum values that is the final result of the segmentation. Afterwards, keeping the matching processes until the six segmentations are calculated completely and the license plate can be reorganized by the recognition results to output the recognized license plate.
21
Sung, Ming-Che, and 宋明哲. "Implementation of License Plate Detection and Recognition System on PDA." Thesis, 2009. http://ndltd.ncl.edu.tw/handle/61703338713751414744.
Full textAbstract:
碩士元智大學
電機工程學系
97
This thesis mainly aims at discussing how to utilize general handheld PDA equipment, detect and recognize to the license plate of the image. The system is developed on Windows Mobile platform. Seeing that the treatment technology of the image is mostly developed on the PC platform, it is unable to make the system miniaturize, melt and save the cost lightly. The purpose is that hopes through the developing tools of Microsoft and cooperates with PDA equipment, easy to use and convenient characteristic transmitted realize that the license plate detection and recognition system. The principle of our system is made up by follow steps, i.e., use PDA capture the picture, pre-processing image, detect and recognize license plate, post-processing image, and finally deliver the result to screen.
22
Ho, Li-Kung, and 何立功. "Implementation of License Plate Detection and Recognition in Embedded System." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/82718990873073525284.
Full textAbstract:
碩士元智大學
電機工程學系
96
A procedure is using Ada-Boosting technique for license plate detection and recognition to develop on hardware platform. According to recent techniques of video processing are almost developed on PC platform, it is heavy, costly, and no space to reduce size of system. However, there are coming more and more embedded systems which are best and best in the future. The goal is making system faster, smaller and cheaper. Our purpose is to develop a license plate system on car and implement it on DSP based hardware system. The principle of our system is made up by follow steps, i.e., capture video, pre-processing image, detection, recognition, post-processing image, and deliver video.
23
Chen, Chia-Hao, and 陳家豪. "Kinect-Based Static/Dynamic License Plate Detection and Recognition System." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/62902829973826480118.
Full textAbstract:
碩士淡江大學
電機工程學系碩士在職專班
100
In this paper, we divide the system into three parts: the extraction of the depth information, the orientation of the license plate and the identification of the words. First, we exact the depth information of the license plate on the motorcycle in front of the Kinect to solve the problem of the complex background. Then we use the algorithm to divide the image of the motorcycles sequentially and abstract the image of one motorcycle to do the following process. After that, we apply the Marker Detection Algorithm with the characteristic of searing the square object to orientate the license plate. The algorithm can not only detect in the real-time but also have the robust resistance to rotation, deform, slope and shelter, etc. Finally, we correct the slope of the retaining license plate and divide the words. We base on the sample to identify the words. The training motorcycle license plates divide into the static and dynamic state. The dynamic state is to detect the driving motorcycle license plate under moving. But, the static state is to detect the parking motorcycle in the roadside under moving. However, our system’s main purpose is the static detection.
24
Chu, Heng-Chin, and 朱恆志. "Application of Embedded License Plate Detection and Recognition for Parking Garages." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/yavrwr.
Full textAbstract:
碩士國立臺北科技大學
自動化科技研究所
101
Transportation is an important topical subject to a country. Under the public demand for convenience, the number of vehicles has continued to increase, which has led to the problems concerning the effective management of vehicle access control and parking fee collection. The automatic license plate recognition technology can solve these problems efficiently. However, the computer-based license plate recognition system has large volume and is power-consuming, thus, is disadvantageous for long-term operation. If the system can be operated on an embedded platform, the problems can thus be solved. Therefore, this study used Texas Instrument DM6437 EVM as the development platform, as well as cameras and LCD display to complete a real-time license plate recognition system. The license plate images are first captured by the cameras installed over or on the side of the parking garage entrance. As the distance and angle between the parking vehicles and the cameras are variable, the images may have angle of tilts. Hough transform is used to detect the inclination. After tilt correction, the character labeling is performed for character segmentation, and the segmented characters are put into the hierarchical character recognition system to identify the results. In order to validate the system feasibility, this study conducted experiments on the vehicles in the parking garages, and proved that the success rate of tilt correction is 88.89% and the overall recognition rate is 81.63%.
25
Wu, Chun-Yen, and 吳俊諺. "Image Motion Object Detection for License Plate and Gesture Recognition System." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/52083337034293954050.
Full textAbstract:
碩士國立雲林科技大學
電子與資訊工程研究所
94
The object detection and recognition is failure on static image of an image recognition system that is caused by the complex background of the scene environment. The characteristic of the dynamic image will not encounter the restrictions of the object and background environment that it interests to the moving object. This thesis utilizes this characteristic to detect dynamic vehicle and gesture. We propose two methods to improve the efficiency of dynamic vehicle and gesture detections and increase the precision rate of the tracking object. First, license plate recognition system consists of image capture, vehicle detection, vehicle tracking, license plate location and character recognition. License Plate recognition system requires more complex computation such that real-time processing is difficult; thus, a FPGA design of License Plate recognition technique with real-time processing is necessary. The simulation results show that the License Plate recognition system has high precision of the license plate segmentation in real-time processing. Second, the gesture tracking and segmentation system is proposed based on dynamic object detection and skin-color model techniques. Our method combines skin-color and hand motion information to detect and segment the gesture in complex background. Experimental results show that the gesture tracking and segmentation technique is able to detect different skin-color in complex background and segment the region of hand for gesture recognition.
26
Yang, Ju-Ting, and 楊茹婷. "Real-Time New License Plate Detection and Recognition on Mobile Platform." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/98689924524073200439.
Full textAbstract:
碩士元智大學
電機工程學系
104
In this thesis, we focus on the new license plate as our target. We built a real-time system for new license plate on android mobile platform. The system is consists of three major parts, which are new license plate localization, character segmentation and license plate number recognition. For license plate localization, Adaboost cascade classifier is used for training with input features obtained by Haar-like features. Cascade classifiers are developed for the purpose of efficient filtering. To recognize license plate efficiently, an open source of Tesseract-ocr provided by Google is used. And we used JavaCV for basic image processing on mobile platform.
27
Bai, Chu-Ping, and 白楚平. "Real-Time License Plate Detection and Recognition Based on LabVIEW System." Thesis, 2017. http://ndltd.ncl.edu.tw/handle/rwp9w6.
Full textAbstract:
碩士國立臺灣海洋大學
機械與機電工程學系
105
In this thesis, LabVIEW system is combined with a webcam to build an Automatic License Plate Recognition System. The image taken by ALPRS is transformed to grey scale image from color space first. Then, the white noise is removed by median filter. Canny edge detection method is used to detect the license plate in the image. Finally, after further morphology process, letters and numbers in the plate image are extracted through template pair technique using pattern matching to achieve license plate recognition.
28
Hsu, Ting-Hsuan, and 許庭瑄. "License plate detection and recognition system based on convolutional neural network." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/phvh67.
Full textAbstract:
碩士義守大學
電子工程學系
107
In recent years, automatic license plate recognition (ALPR) system is applied in some traffic-related applications based on deep learning. However, most ALPR systems capture a mostly frontal view of the vehicle and license plate (LP) to obtain high LP recognition rates. This is because the traditional ALPR cannot capture the correct area of oblique LP which results in an error in character recognition or missing characters. As a result, the traditional ALPR will largely reduce the accuracy of recognition for oblique LP. Recently, Silva et al. [12] proposed a warped planar object detection (WPOD) based on convolutional neural network (CNN) to overcome the oblique views of LP. In order to achieve an ALPR system of high accuracy, they divided ALPR into three stages. The first stage is to locate vehicles through YOLOv2 [9]. And then, the second stage locates the oblique LPs and allows a rectification of the LPs area to a rectangle which resembles a frontal view through the WPOD network. Finally, the rectified LPs are fed to an optical character recognition (OCR) in the third stage. Although the WPOD network can achieve the location and rectification of LPs, the loss function of WPOD render the confidence parameter due to high computational complexity. This also leads to WPOD network cannot locate the optimal LP bounding box. In order to further improve the accuracy of ALPR system, we proposed a modified WPOD network using a complete loss function. The proposed method first develop a simple intersection over union (IOU) algorithm to speed up the calculating process of confidence. Therefore, the modified WPOD network can obtain higher LP recognition rate since it considers the confidence parameter in loss function. In this thesis, the four-vertex coordinates of the label bounding box and prediction bounding box of oblique LP are used to generate two rectangular boxes, and then a simple IOU algorithm is used to fast calculate the approximate value of IOU. As a result, a more exact loss function can be finished. In order to compare the performance of LP recognition rate, we train and test the WPOD and the proposed modified WPOD in databases including OpenALPR EU, BR [19], and AOLP RP [17]. Simulation results show that the proposed ALPR system can obtain higher score ratios than those of Silva’s method. And the proposed system can arrive a high accuracy of LP recognition about 95% on an average. In addition, the proposed system also can achieve higher recognition rate about 1% when compared to the Silva’s ALPR system.
29
Liang, Hui-Chen, and 梁輝城. "The Study of the Detection of Non-fixed Distance Car License Plate Position." Thesis, 2010. http://ndltd.ncl.edu.tw/handle/73130261012839706980.
Full textAbstract:
碩士中華技術學院
電子工程研究所碩士班
98
The study is to detect the position of the car license plates of vehicles in the picture obtain from a consumer camera with the VGA resolution. It can be applied to the images of license plates of vehicles captured from the different locations, distances, and angles to detect the locations of license plates via the proposed process. Due to the image to be used must be obtained to have the different sizes of license plates of vehicles and the complicated background, the images of license plates of vehicles are captured from the different locations, distances, and angles. The locations of license plates are frequently incorrectly detected if a single set of parameters is only used to detect the positions of the license plates of vehicles. Therefore, the parameters of various set of parameters are designed to detect the locations of license plates. The parameters to be adjusted include the parameters of various rectangular windows, the parameters of median filtering and vertical edge enhancement filter, and the best threshold to perform the binarization process. Once with the RGB color space separation, we can locate the vehicle license plate position preliminarily. Meanwhile, put the template and image of license plate matching in advance to the secondary accurate position detection for the license plate. The research is to use aforementioned methods to do the experimental test with more than 1000 vehicle license plate images. The results showed that using multiple sets of parameters to locate license plates has a higher successful rate.
30
Wang, Shih-han, and 王詩涵. "Automatic Multiple Vehicles’ License Plate Detection and Recognition System in Complex Environment." Thesis, 2013. http://ndltd.ncl.edu.tw/handle/25731700373072197341.
Full textAbstract:
碩士國立臺灣科技大學
電子工程系
101
This system is specially built for Taiwan’s environment. It has 4 parts: image preprocessing, multiple license-plate detection, normalization and recognition. In image preprocessing, this system takes the benefit of license plate’s color feature then come out a color differential edge detection (CDED) algorithm combines with horizontal Sobel mask, which can erase 62.5% of uninteresting pixels. In detection part, finding edge pixel of character feature then using morphology and density filter extracts the real license plate region. This method can find all license plates region in one scan and has 96.41% succession rate therefore consuming time is 168ms. Normalization part has to solve license plate’s rotation and warp to increase recognition succession rate. This system uses characters in license-plate to estimation rotation angle, and can normalize angle of 0°~36°. In recognition part, projection segmentation combines with segmentation region estimation method can segment sticky characters and increase recognition rate. Finally this system uses probability templates as OCR which has weight for each pixel for every character, and it is fast and has 97.89% succession rate. Of all, this System’s recognition rate is 89.16% and consuming time is 241ms. This paper also collects nowadays LPR related papers, and this system has less restricts and higher succession rate than those.
31
HUANG, ZIH-JIA, and 黃子嘉. "A Research and Implementation on Image-based License Plate Detection and Recognition." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/4jejd7.
Full textAbstract:
碩士國立高雄應用科技大學
資訊工程系
105
Image processing is a technology of image machining, analysis and treatment for the objects accessing the customized range to obtain required information. For example, with image processing, the game industry may make players synchronously interact with games. Through processing action images of players, corresponding actions can be made by games to enhance the interest thereof. Furthermore, the main transportation in Taiwan is motor vehicles, and the number of motor vehicles has reached up to 21412810 in accordance with Ministry of Transportation in 2016. Accordingly, when the problems are caused by such a huge number of motor vehicles, owners must be found by the license plates. Since license plates are as unique as identification cards, it is important to recognize the license plates for camera image on the road. Therefore, the present research conducts the detection and recognition of the license plates at the gate and parking shed of NKNU. Specifically, the vehicles in a range can be got by technology of image processing for the real-time image of IPCAM, and the position and the actual information of license plates can be detected and recognized . By transferring information to database and matching therewith, whether the plate number applies for the licenses can be confirmed. If not, the warning light is flashed by issuing the command to inform the vehicle to stop. As such, it can not only improve the working efficiency but save more personnel cost.
32
Lin, Yu-You, and 林俞佑. "Vehicle License Plate Detection and Recognition for Roadside Parking via Deep Learning." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/5u5ar6.
Full textAbstract:
碩士臺北市立大學
資訊科學系
107
According to the Ministry of transportation and communications R.O.C. statistics, Taiwan’s registered motor vehicles has been a steady increase, and more and more. Therefore, these work related to vehicles registration require a considerable amount of manpower. For example, in order to manage roadside parking vehicles, the government outsource the charge work to private companies. And the work of the roadside parking toll collector often shuttles on the road with a high traffic flow. Moreover, to identify the license plate and then manually enter the PDA system, the work is not only dangerous but also actually very inefficient. If there is an intelligence license plate recognition system, the parking toll collector can use mobile device or smart glasses as tool to help their jobs. The parking toll collector just slowly move from the roadside parking space, they use the device to scanning the roadside parking space, and the system can automatically detect and recognize the license plate. In this way, the Government can effectively reduce unnecessary manpower, and can also improve the efficiency of roadside parking toll collector, and enhance the safety of work. In addition, it can also be used in police service and some license plate registration related work. Therefore, this study proposes a combination of methods, uses deep learning model to detect the license plate in the scene, after that to recognize the license plate number. Take Taiwan's vehicle license plates as examples and collect roadside parking license plates as sample data. Through experiments, the method identified the roadside parking license plate has a good identification effect, and it is better than the commercial software. The overall accuracy rate is 80.6%.
33
Chen, Yan-Bin, and 陳彥賓. "A STUDY FOR APPLYING REVERSE LOGIC ARTIFICIAL NEURAL NETWROK IN CAR LICENSE PLATE CHARACTER RECOGNITION." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/79243523298623853663.
Full textAbstract:
碩士大同大學
資訊經營學系(所)
93
Over the recent years, as the vehicle license plate recognition system technology continues to breakthrough, the actually potential applications are also on the rise. The common examples are detecting stolen cars ticketing traffic violations, and the toll collecting systems of man-free parking lots. Yet in practical implementation, a main reason that the license plate recognition system is unable to rapidly and expeditiously become prevalent lies in how the recognition technology leaves room to be improved upon, such as the license plate positioning ratio, character recognition ratio and so forth. As a result, the study intends to focus on this area in anticipation that the recognition accuracy of the license plate recognition system can be further enhanced. This study utilizes back-propagation neural network (BPNN) as the recognition system tool. However, the back-propagation neural network does have a few deficiencies that need to be improved upon, such as the issues of a slow learning speed in the training process are prone to lead to partial minimum values that are difficult to converge, and the need to retrain an enormous volume of data whenever new training samples are added or deleted, which tends to crate significant obstacles in practical implementation. In light of that, this study intends to focus on the conventional back-propagation neural network’s learning paradigm and network framework combining normalization, parallel dispersed and reverses logical thinking concepts to propose a reverse logical artificial neural network (RLANN). The RLANN is intended to improve some of the back-propagation neural network’s problems of a slow learning speed in training that makes normalization difficult, and the requirement for a fixed amount of training samples. By the way, a practical validation of the license plate character recognition system is conducted to validate the performance improvement. The study has only adopted 45 character samples to train the reverse logic artificial neural network, and employed 200 license plates as test samples. The overall license plate recognition ratio reaching 91%, and the overall character recognition ratios of 98.5% have validated the results to be satisfactory.
34
Chung, Meng-Liang, and 鐘孟良. "The Study of Vehicle Detection Technologies for the Application of License Plate Recognition Systems." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/v3nwbn.
Full textAbstract:
博士國立交通大學
電控工程研究所
104
With the development for “Internet of Things (IoT)” technology, Internet of Vehicle technology has become mature. Such systems, nevertheless, focus only on individual vehicle’s driving conditions and often ignore that of the surrounding vehicles, thus making “Internet of Vehicle (IoV)” technology significantly crucial. This paper proposes two systems to obtain the big data analytics of moving vehicles, and to construct a safety and wisdom road system by IoV. The first proposed system utilizes the location of the vehicle to trigger the camera to capture an image of the vehicle. This not only achieves the goal of detecting an image in real time, but as well obtains the best image with the license plate as a triggering image. The proposed algorithm does not require additional hardware, facilitating the precise retrieval of an image that both contains a vehicle and represents the best image from a series of images for recognizing the characters on license plates. This paper proposes another effective real-time detection algorithm that uses two low-cost compact Complementary Metal-Oxide Semiconductor (CMOS) cameras as vision sensors to detect front vehicles and measure distances. This low-cost vision sensor is an asynchronous vision system since a custom-made binocular is used to capture pairs of images via a low-cost grabber card and the left and right images have a slight time difference. Nevertheless, the proposed detection algorithm, which comprises four modules—namely, image preprocessing, vehicle detection, detected vehicle tracking, and distance measurement—is unrestrained by the limitations of the epipolar constraints for synchronous vision systems, and also enables the system to overcome the asynchronous detection problem affecting conventional low-cost vision systems. The results of long-term performance tests conducted on highways and urban and country roads confirm that the proposed system can successfully detect the distance of the front vehicle. The proposed algorithms significantly reduce the cost of hardware and are much easier and cheaper to maintain than the traditional detection methods. An extensive vehicle-detection test demonstrates that the proposed algorithm is reliable and accurate. The analysis of the big data obtained by the proposed algorithms can provide the information of IoV and construct a safety and wisdom road system.
35
Hsieh, Chengta, and 謝正達. "License Plate Detection and Recognition Using a Dual-Camera Module in a Large space." Thesis, 2007. http://ndltd.ncl.edu.tw/handle/17867602066071188902.
Full textAbstract:
碩士中華大學
資訊工程學系碩士班
95
In this study, a calibrated dual-camera device, a fixed camera and a pan-tilt-zoom camera, is setup to monitor moving vehicles in an open space. This device not only tracks multiple targets but also gets the license plate images with high quality. Next, a simply image processing method is used to locate the license plate and a back-propagation Neural Network is designed to be a character classifier for efficiently recognizing the alphabets on them. Two working environments were setup at the entrance of university and a pedestrian-only region in campus. Some experimental results are given to show the validity of the proposed approach.
36
Wong, Cing-Jhang, and 翁慶彰. "Video-Based License Plate Recognition for Software-Assisted Parking Violation Detection and Citation Issuing." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/fp24j2.
Full textAbstract:
碩士國立臺灣海洋大學
電機工程學系
106
Illegal roadside parking not only can obstruct the normal traffic, but also can have negative impact on the safety of pedestrians and other vehicles. Among the many different patterns of illegal roadside parking, reverse parking is particularly life-threatening because it involves dangerous wrong-way driving. In addition, vehicles without license plates can sometimes be found on the roadside. These vehicles can cause not only traffic problems, but also issues related to sanitation and public health, if left undealt with for a prolonged period. The objective of this thesis is to develop a software tool that is capable of recognizing vehicle license plates from videos taken by patrol cars to assist citation of illegal roadside parking. In particular, reverse parking and vehicles with no license plates will be identified automatically. When a license plate cannot be recognized due to obstruction or smearing, a warning will be issued, so further actions can be taken by the responsible parties. The software implementation has been tested using 39 field videos that contains 335 vehicles. Experimental results show that the software performs better when the light condition is relatively even, such as when the sky is cloudy. As expected, dented or smeared license plates can cause difficulties in recognition. Under various weather and lighting conditions, the total recognition accuracy for the current implementation is about 85%.
37
Shun-bin, Jhuang, and 莊順斌. "A Product Appearance Inspection System Based On Modified Sobel Edge Detection And Density-Based Clustering Algorithm Taking Car-License Recognition As Example." Thesis, 2005. http://ndltd.ncl.edu.tw/handle/21312940352322646640.
Full textAbstract:
碩士國立屏東科技大學
資訊管理系
93
License Plate Recognition（LPR）is a complex and important research issue. In this issue, there are many techniques and theories applied for solving this problem. For an image recognition system , edge detection is an extremely significant pre-processing step and thus we design a novel edge detection approach from modifying traditional Sobel edge detection method. The modified Sobel edge detection is suitable for processing images with colorful objects and backgrounds. Therefore, we develop a product appearance inspection system based on it. Moreover, we adopt density-based clustering algorithm to locate license plates. Notably, the modified Sobel edge detection and density-based clustering are the two main stages in this system. The proposed system structure makes our real-time vehicle license recognition system works quickly and efficiently. Through our experiments, this system has some obvious advantages as follows: (1) It operates with less resources than other structures. (2) It is easy to implement applications based on this framework.