|
|
Detection, Tracking and Classification of Vehicles
Vopálenský, Radek ; Herout, Adam (referee) ; Juránek, Roman (advisor)
The aim of this master thesis is to design and implementation in language C++ a system for the detection, tracking and classification of vehicles from streams or records from traffic cameras. The system runs on the platform Robot Operating System and uses the OpenCV, FFmpeg, TensorFlow and Keras libraries. For detection is used cascade classifier, for tracking Kalman filter and for classification of the convolutional neural network. Success rate for detection is 91.93 %, tracking 81.94 % and classification 63.72 %. This system is part of a comprehensive system, that can moreover calibrate video and measure of vehicles speed. The resulting system can be used for traffic analysis.
|
|
|
Neural Network Number Recognition
Doupovec, Zdeněk ; Juránek, Roman (referee) ; Šilhavá, Jana (advisor)
This work describes the basic concepts and principles in the field of neural networks. Closer then deals with the problem of multilayer perceptron networks, namely back-propagation method. There are analyzed the advantages and disadvantages of these methods, the proposal of possible digits recognition system using back-propagation. The aim is to obtain concrete results from the program whitch is able to recognize numbers.
|
|
|
RoboAuto - Detection of Moving Objects
Štibinger, Petr ; Žák, Pavel (referee) ; Juránek, Roman (advisor)
In this work is possible to find out something about detection of moving objects from moving camera. Method that was used is MTI (moving target indication), which is used for detection from air and can be applied on ground vehicles. Introduced will be work with SURF (for detection of interesting points in image) and histograms (for detection moving objects).
|
|
|
Automatic Rectification of Photos
Burkalo, Boris ; Herout, Adam (referee) ; Juránek, Roman (advisor)
The aim of this thesis is to propose a method for automatic rectification of larger sets of photos. This thesis is first aimed at describing the problem which is posed by rectification of photos and the description of individual steps that need to be taken to successfully rectify an input photograph. After that the thesis describes individual methods for line and vanishing point detection. Next the proposed method for automatic rectificator is implemented and ultimately a statistic of used methods is created.
|
|
|
Merging of Different Exposured Photographs, HDR
Šmida, Vladimír ; Juránek, Roman (referee) ; Seeman, Michal (advisor)
High Dynamic Range Imaging (HDRI) is an emerging field in image postprocessing with non-trivial relations to many different areas, such as image synthesis, computer vision, video and image processing, digital photography, special effects among others. This bachelor's thesis deals with the usage of HDRI in digital photography. Conventional digital cameras are not able to record the full tonal range of the scene. Therefore, if we want to capture a high dynamic range scene, by setting the exposure time we have to choose only a small range of tones which will be well depicted. The rest of the photograph will loose important detail. The method presented and implemented in the scope of this thesis enables us to create HDR picture with realistic values of tones out of several differently exposured photographs.
|
|
|
Biologically Inspired Methods of Object Recognition
Truhlář, Martin ; Hradiš, Michal (referee) ; Juránek, Roman (advisor)
This document describes a method for biologically inspired pattern recognition. Furthermore, it explains the process of image processing and various stages of information extraction for classification. Support Vector Machine method is used for classification, but there are other classification methods explained. It explains how to test and work with the method itself. Results for each model set of classifiers and their advantages and disadvantages are summarized in the conclusion.
|
|
|
Face Reader
Bučko, Peter ; Juránek, Roman (referee) ; Beran, Vítězslav (advisor)
This thesis deals with computer face recognition. Methods of Components Analysis (PCA), Linear Discriminant Analysis (LDA) and Elastic Bunch Graph Matching (EBGM) are described here. Aim of this thesis is creation of a demonstration aplication for a face recognition. Moreover I test PCA and LDA methods to find out, how accurate it can be and how can be affected by changing of parameters, such as size of a database and picture count per person.
|
|
|
Rendering Biliard Balls Using Distributed Ray Tracing
Krivda, Marian ; Juránek, Roman (referee) ; Herout, Adam (advisor)
This thesis is concerned in the method of realistic rendering using a distributed raytracing. This method simulates various visual effects and generates high realistic 2D images. The work analyses the problem and explains principles of solution related to this technique. There is also descriprion of the method of simple reytracing which provides a basis for the distributed raytracing. A part of work is specialized for optimalization of distributed raytracing.
|
|
|
Head Pose Estimation from Photography
Szentandrási, Tamás ; Juránek, Roman (referee) ; Behúň, Kamil (advisor)
The aim of this work is to propose and implement a head pose estimation system using Random Forests, with photographic input. This work uses a version on Random Forests with using Fisher Linear Discriminant Analysis for node splits. It also uses a number of tools to achieve this goal, the OpenCV library, SQLiteCpp library, and the aflw database. This work created an accurate head pose estimation system reusable for future Computer Vision tasks.
|
|
|
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.
|
guest ::
