|
|
Automatic Video Colorization
Mikeska, Tomáš ; Kolář, Martin (referee) ; Hradiš, Michal (advisor)
This thesis deals with fully automatic colorization of video and images using convolutional neural networks. It summarizes existing approaches, architectures and several objective functions. In this thesis several neural networks with different objective functions are implemented and trained for automatic colorization. The best of them is able to colorize large diversity of scenes and is sufficient to produce coherently colorized video.
|
|
|
Thermal Desorption Spectroscopy (TDS) and its Application for Research of Surface Processes
Potoček, Michal ; Čech, Vladimír (referee) ; Pavlík, Jaroslav (referee) ; Dub, Petr (advisor)
ermal desorption spectroscopy (TDS) is a common method for surface analysis of adsorbed molecules. In chapter 1 the work deals with the theoretical background of this method and shows the principles of a desorption process influenced by subsurface diffusion. Chapter 2 first shows application of TDS for detection of surface molecules and determination of binding energy.Experiments were mainly focused on ditermination of surface adsorbents and impurities on Si wafers. The second part of chapter 2 describes desorption of atoms of a Ga layer on Si surface and their subsurface diffusion. A Ga diffusion process was also observed by with secondary ion mass spectrometry (SIMS) and numerically simulated.
|
|
|
Learning the Face Behind a Voice
Zubalík, Petr ; Mošner, Ladislav (referee) ; Plchot, Oldřich (advisor)
The main goal of this thesis is to design and implement a system that will be able to generate a face based on the speech of a given person. This problem is solved using a system composed of three convolutional neural network models. The first one is based on the ResNet architecture and is used to extract features from speech recordings. The second model is a fully convolutional neural network which converts the extracted features into the styles which form a base for the final facial image. These styles are then passed as an input to the StyleGAN generator, which creates the resulting face. The proposed system is implemented in the Python programming language using the PyTorch framework. The last chapter of the thesis discusses some of the most significant experiments performed to fine-tune and test the developed system.
|
|
|
Inverter for electric supercharger with GaN transistors
Galia, Jan ; Pazdera, Ivo (referee) ; Martiš, Jan (advisor)
This master’s thesis deals with the design and realization of a functional sample power inverter for an electric compressor, which is used in hybrid cars. The electric compressor powered by the inverter is E-compressor by Garrett Advancing Motion. An inverter will be using modern High Electron Mobility Transistors which are based on gallium nitride (GaN). The purpose of this thesis is to find if GaN transistors can be used in E-boosting application.
|
|
|
GaN deposition on a tungsten substrate
Pikna, Štěpán ; Piastek, Jakub (referee) ; Čalkovský, Vojtěch (advisor)
This bachelor thesis is focused on deposition of GaN nanocrystals on the etched tungsten tips. Motivation was to prepare these GaN structures on the Schottky cathode made by company ThermoFisher Scientific and measure its field emission. In the theoretical part of the thesis GaN and tungsten field emission properties are introduced. The experimental part begins with tungsten tip etching optimalization, where the right values for best tips are temperature 20 °C, depth of the tip 2,5 mm and solution NaOH used. Further the gallium structures were prepared on these tips using molecular beam epitaxy (MBE). The right temperature to prepare GaN nanocrystals was determined as 200 °C. The deposition of gallium was set to 2 hours and following nitridation was 3 hours. Finally, the field emission from GaN prepared on copper foil with graphene was measured and compared with other experiments.
|
|
| |
|
|
Generative Neural Networks for Handwritten Text
Ševčík, Pavel ; Dobeš, Petr (referee) ; Hradiš, Michal (advisor)
The aim of this study was to create a generative neural network for handwritten text lines. The model produces variable-sized images of handwritten text lines based on the expected style. The proposed method exceeds existing models in the image quality and can be used to generate both individual words and entire lines of handwritten text. It combines the use of the attention mechanism to extract the features for each character from the text query and their arranging on the line by inserting spaces between them. The new approach allows more granular control of the symbol positions on the line, which leads to smoother style interpolations. In contrast to the previous approach, the proposed method uses the Gaussian filter to spread the individual symbols features to the surrounding area. This approach also allows to train the model for symbols position predictions using the adversarial loss (GAN). In addition, annotations of symbol horizontal positions on the lines of the IAM dataset of handwritten text have been created.
|
|
|
Detection and Quality Improvement of Face Objects in Low-Quality Source Images
Šoltis, Richard ; Tinka, Jan (referee) ; Drahanský, Martin (advisor)
The aim of this thesis was to construct an algorithm for the detection of human face from poor quality source images and subsequently improving the image of human face. The result of the work is an application with a graphical interface which detects human face objects from the input images and then improves these inherited faces from the point of quality and size. When creating the application, current techniques and algorithms such as neuron networks were used. They formed the basis for detection and image improvement, S3FD detection and last but not least the GAN network to improve the image. Part of the thesis is testing the individual parts of the application in predefined scenarios as well as testing a comprehensive run application.
|
|
|
Algorithms for improving the detection of selected cardiac arrhythmias
Šandová, Hana ; Ředina, Richard (referee) ; Novotná, Petra (advisor)
The work deals with the generation of ECG arrhythmias that are underrepresented in databases. The theoretical part of the thesis is devoted to a literature search of academic publications that deal with the classification of arrhythmia by using deep learning and data augmentation metod for ECG. The practical part of the thesis deals with noise generator, because adding noise to signals could make the dataset richer. Functions for augmentation of atrial flutter and 3rd and 2nd atrioventricular block were created. It has been tried generation of 2nd atrioventricular block using generative adversarial networks (GAN). Deep learning-based ECG classifiers were used for evaluating the efficiency of the proposed technique in generating synthetic ECG data.
|
|
|
Fusion of Radar and Visual Data for Remote Sensing
Strych, Tomáš ; Beran, Vítězslav (referee) ; Kolář, Martin (advisor)
Cieľom práce je vygenerovať satelitný optický snímok v prípade jeho nedostupnosti. Takýto snímok je vygenerovaný z aktuálneho radarového snímku a za pomoci radarových a optických snímkov z minulosti. Zameranie práce cieli na poľnohospodárstvo, kde sa na analýzu dát používajú rôzne vegetačné indexy. Pre zjednodušenie problematiky je práca zameraná len na optický snímok zobrazujúci NDVI. Boli vytvorené 4 dátové sady, pre prvé tri ročné obdobia a~štvrtý, ktorý ich spája. Pre riešenie problému preloženia obrázku z jedného na druhý bol použitý model Pix2Pix-cGAN. Výsledky práce zobrazujú rozdiely pri použití dátových sád, rozličného množstva a~typu použitých snímkov, tak ako aj intervalu medzi snímkami. Daným výskumom bolo zistené, že sieť je schopná vytvárať reálne uveriteľné obrázky s validnými numerickými hodnotami, avšak má problém správne využiť informáciu o radarovej zmene, ktorá je dôležitá pre ohodnotenie vývoja rastlín práve v prípade nedostupnosti optického snímku. Táto práca a~jej výsledky sú jedinečné vďaka naprieč Európou geograficky rozmanitej dátovej sade a vďaka zameraniu na agrikultúru, a to bez ohľadu na typ plodín.
|
guest ::
