|
|
Steps Towards Improvements of Computer Vision Methods for Traffic Analysis
Špaňhel, Jakub ; Sablatnig, Robert (referee) ; Šikudová, Elena (referee) ; Herout, Adam (advisor)
Rostoucí urbanizace a zvyšující se počet vozidel na silnicích přetěžují tradiční systémy řízení dopravy na hranici jejich možností. Řešení nabízejí inteligentní dopravní systémy (ITS), které využívají pokročilé technologie ke zvýšení plynulosti a bezpečnosti dopravy. Zásadní oblastí, kterou je třeba zlepšit, však zůstává robustnost metod počítačového vidění v rámci ITS, které jsou nezbytné pro analýzu dopravy. Tato práce přispívá k této oblasti, konkrétně se zaměřuje na přesné (fine-grained) rozpoznávání vozidel, reidentifikaci vozidel, rozpoznávání registračních značek a monokulární měření rychlosti vozidel. Bylo představeno několik nových datových sad, vysoce ceněných výzkumnou komunitou, které rozšiřují hodnocení a zkoumání v každé z výše uvedených oblastí. Hlavní přínosy lze shrnout následovně: Nové technicky augmentace pro přesné rozpoznávání vozidel & rozšíření dříve publikované datové sady. Nová metoda agregace vizuálních znaků pro re-identifikaci vozidel & datová sada. Inovativní přístup k rozpoznávání registračních značek pomocí zarovnání registrační značky a holistického rozpoznávání & tři publikované datové sady. Největší datová sada pro měření rychlosti vozidel & stanovení výchozího vyhodnocení s dostupnými metodami vizuálního meření rychlosti. Klíčová zjištění této práce prokazují významné zvýšení přesnosti, účinnosti a robustnosti metod počítačového vidění aplikovaných na analýzu dopravy. Přínosy tohoto výzkumu byly oceněny na nejvýznamnějších konferencích a v časopisech v oblasti ITS a stanovují nové standardy pro budoucí práci. Tím, že tato práce posunula současný stav ITS a přispěla cennými zdroji pro probíhající výzkum, představuje zásadní krok směrem k udržitelnějším, efektivnějším a inteligentnějším dopravním systémům. Má důsledky pro řízení dopravy a širší společenský cíl vytvořit citlivější a přizpůsobivější městské prostředí.
|
|
|
License Plate Recognition
Mrhač, Ondřej ; Sochor, Jakub (referee) ; Navrátil, Jan (advisor)
This thesis talks about problematics of licence plate detection, licence plate recognitionand my implementation for device i.MX 6 Series of NXP semiconductors s.r.o company. Model program for licence plate detection and recognition is written with help of OpenCV library and engine Tesseract and it’s successfully put into operation on this device. Afterwards program was measured his runtime on PC and i.MX6 Series device and those measurements were compared. At the end of this thesis were found the most demanding parts of the program. Future changes and improvements were proposed.
|
|
|
License Plate Recognition
Tilňak, Tomáš ; Juránek, Roman (referee) ; Herout, Adam (advisor)
This thesis talks about a license plate recognition problematics and my implementation of license plate recognition program. At first I introduce a format of license plates in Czech republic. Next chapter is about existing solutions for each phase of license plate recognition according to the selected scientific articles. The main part of this thesis is about design and implementation of license plate recognition program. I also introduce libraries I used in implementation. Necessary part in software development is testing, which has also its own chapter. In the final part there is a review of results and proposals for future changes.
|
|
|
Holistic License Plate Recognition Based on Convolution Neural Networks
Morbitzer, Dušan ; Juránek, Roman (referee) ; Špaňhel, Jakub (advisor)
The goal of this work is to create a model of neural network for holistic recognition of license plates, focused on accuracy and shortening of the learning process. The model was implemented as a union of convolutional neural network for extraction of deep features of a plate and Bidirectional LSTM with CTC. The trained model was compared to another implementation using a holistic approach, that was trained on the same dataset. My design of the network achieved better results in recognition on a dataset, which is different from the training one, with an error rate of 8.3 %.
|
|
|
Re-Identification of Vehicles by License Plate Recognition
Špaňhel, Jakub ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
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.
|
|
|
Image processing using Android device
Hortai, František ; Petyovský, Petr (referee) ; Honec, Peter (advisor)
This thesis describes the design and workflow of creating an image processing application in Android system, and what are the possibilities in choosing development environment and how to implement them. Then I am writing about my solutions of creating applications, graphical user interface and an interface for Android. I am describing my approach in the design and functionality of the application, communication with the camera, storing and retrieving data. Further I explain which algorithms were implemented for image processing and image evaluation. Product of this thesis is a functioning application that allows to its user to capture images and video stream. The algorithm evaluates the entering data and shows the location of the number plate. The application also allows recognizing texts and numbers from images. There are other various practical features and options implemented within the application.
|
|
|
On-Board License Plate Detection and Recognition
Tomovič, Martin ; Sochor, Jakub (referee) ; Špaňhel, Jakub (advisor)
This Bachelor's thesis aims to create an aplication for detection and recognition of license plates suitable for real-time processing. The work contains analysis of available methods. Part of the work is focused on present form of licence plates in Czech Republic. As a result of work, new data set was created and computer application was implemented. The application uses existing libraries designed for computer vision and machine learning with main purpose to detect and recognize licence plates from video. Detection is realized with help of cascade classifier, and recognition by Perceptron neural network. Final chapter subsequently contains evaluation of success rate of implemented solution.
|
|
|
Holistic License Plate Recognition Based on Convolution Neural Networks
Morbitzer, Dušan ; Juránek, Roman (referee) ; Špaňhel, Jakub (advisor)
The goal of this work is to create a model of neural network for holistic recognition of license plates, focused on accuracy and shortening of the learning process. The model was implemented as a union of convolutional neural network for extraction of deep features of a plate and Bidirectional LSTM with CTC. The trained model was compared to another implementation using a holistic approach, that was trained on the same dataset. My design of the network achieved better results in recognition on a dataset, which is different from the training one, with an error rate of 8.3 %.
|
|
|
License Plate Recognition
Mrhač, Ondřej ; Sochor, Jakub (referee) ; Navrátil, Jan (advisor)
This thesis talks about problematics of licence plate detection, licence plate recognitionand my implementation for device i.MX 6 Series of NXP semiconductors s.r.o company. Model program for licence plate detection and recognition is written with help of OpenCV library and engine Tesseract and it’s successfully put into operation on this device. Afterwards program was measured his runtime on PC and i.MX6 Series device and those measurements were compared. At the end of this thesis were found the most demanding parts of the program. Future changes and improvements were proposed.
|
|
|
Re-Identification of Vehicles by License Plate Recognition
Špaňhel, Jakub ; Juránková, Markéta (referee) ; Herout, Adam (advisor)
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.
|
guest ::
