|
|
Detekce začátku a konce komplexu QRS s využitím hlubokého učení
Müller, Jakub ; Šaclová, Lucie (referee) ; Smíšek, Radovan (advisor)
ECG measurement isan essential diagnostic tool for cardiac health, and automation of its analysis can aid to our healthcare to relieve staff workload or improve the quality of automated diagnostics from wearable devices. This work focuses specifically on the QRS complex in the ECG signal, with the main goal of using deep learning methods to detect its onset and offset. In the theoretical introduction, the reader is introduced to the origin of the QRS complex and ECG measurements, artificial neural networks and deep learning. Modified architecture U-Net for 1D signals was chosen to implement the actual method. Data were extracted from five publicly available databases and preprocessed in Matlab. This was followed by moving to the Python environment where parts of the model were implemented using the TensorFlow and Keras libraries, subsequent training, testing of the model and evaluation of the results.
|
|
|
HDR Tone Mapping for Low-Light Conditions
Macejka, Lukáš ; Polášek, Tomáš (referee) ; Čadík, Martin (advisor)
This bachelor's thesis focuses on the development and implementation of three specific tone mapping methods into the existing TMS system. The tone mapping process involves converting images with high dynamic range to images with low dynamic range. Its importance lies primarily in its ability to display images originally with high dynamic range on devices that only support standard dynamic resolution. The work thoroughly explains the basic principles of digital imaging and the methods used in this process. Each of the implemented methods is described in detail, and their results are subsequently compared in terms of the quality of the resulting images. The goal of the work is not only to adapt the image according to specific requirements but also to integrate new techniques into the system, which should lead to improved functionality and overall visual quality of the image outputs.
|
|
|
Guided Reinforcement Learning for Motor Skills
Karabelly, Jozef ; Herout, Adam (referee) ; Hradiš, Michal (advisor)
Cieľom tejto práce je prezentovať prehľad aktuálneho výskumu v oblasti posilovaného učenia pohybu s predlohou a identifikovať potenciálne smery výskumu. Okrem toho práca predstavuje vylepšenú metódu učenia fyzikálne simulovaných animácií postáv založenú na aktuálnych metódach. Predtrénovaný model ukazuje potenciál lepších výsledkov na rôznych nových úlohách. Vlastný dataset bol nazbieraný pre účely pretrénovania modelu predstaveného v tejto práci. Na základe výsledkov z vykonaných experimentov sú odprezentované možné budúce vylepšenia a smery výskumu.
|
|
|
Genetic Programming with Memory for Symbolic Regression
Jůza, Tadeáš ; Bidlo, Michal (referee) ; Sekanina, Lukáš (advisor)
The purpose of this thesis is to evaluate the possibility of extending genetic programming with memory for solving symbolic regression problems. Furthermore, a set of problems for testing the quality of such solutions is developed. The thesis proposes a practical application of such an extension to reduce the energy consumption of loading weights of convolutional neural networks. Instead of retrieving all the weights of the network from external memory, only a small percentage of the weights is retrieved and the remaining ones are generated using the evolved expression. This method was primarily evaluated on reducing the set of weights of convolutional layers of a small convolutional neural network classifying the MNIST dataset. Furthermore, the possibility of generating weights was also tested on other convolutional neural networks solving more complex classification problems. The proposed method has delivered interesting tradeoffs between the classification accuracy and weight memory size.
|
|
|
Neural network inference on the ZYNQ
Masár, Filip ; Bidlo, Michal (referee) ; Mrázek, Vojtěch (advisor)
Neural networks are becoming increasingly popular. Inference is now performed not only on high-end GPUs, but also on low-power embedded systems. This bachelor’s thesis explores ways to test fault tolerance on the hardware accelerator of neural networks. It propose the use of FPGAs to increase the performance of fault tolerance experiments. To achieve this goal, an open-source accelerator NVDLA was used and modified to support fault injection. Furthermore, an analysis of the fault tolerance of ResNet-18 is presented to demonstrate the proposed solution.
|
|
|
AI-based classification of RF signals
Turák, Samuel ; Ulovec, Karel (referee) ; Polák, Ladislav (advisor)
Táto práca sa zameriava na klasifikáciu rádiofrekvenčných (RF) signálov založenú na hlbokom učení. Pre tento účeľ, tri neuronové siete sú vybrané a prezentované: Konvolučná Neurónová Sieť (CNN), Sieť s Bránovými Rekurentnými Jednotkami (GRU), Konvolučná Hlboká Neurónová Sieť s Bránami (CGDNN). Všetky sú trénované a vyhodnotené na viacerých datasetoch, ovplyvnené rôznymi RF rušeniami, pre klasifikáciu rôznych bezdrátových štandardov. Signály v jednotlivých datasetoch boli vytvorené pomocou aplikácie Wireless Waveform Generator v programu MATLAB. Jeden verejne dostupný dataset na klasifikáciu modulácie je takisto testovaný na modeloch. Použité prístupy k predspracovaniu dát, tréningu modelov a vyhodnoteniu modelov sú implementované v programovacom prostredí Python s využitím knižníc ako Scikit-learn a Keras. \mbox{Získané výsledky} sú prehľadne prezentované a diskutované.
|
|
| |
|
|
Design of signals for nonlinear system modeling
Kuba, Michael ; Ištvánek, Matěj (referee) ; Miklánek, Štěpán (advisor)
This Master’s thesis is focused on training signals for nonlinear system modeling using deep learning. A theoretical introduction to the problem is described, including an initial description of signals and nonlinear distortion audio effects. The goal of the thesis is to design a set of artificial training signals for creating models of distortion guitar effects or tube guitar amplifiers. The designed set of artificial training signals is then processed, utilizing guitar and bass guitar effects, and using a recurrent neural network their models are trained. The quality of the resulting models is afterwards compared with the quality of the models trained with the help of a reference training signal, composed of electric guitar recordings, and signals from commercially available devices. The comparison is carried out in accordance to objective metrics and with subjective evaluation by the MUSHRA listening test.
|
|
|
Automated creation of deep neural network models for image classification
DOHNAL, Patrik
The aim of the thesis is to design and implement a system that can automatically create deep neural networks (DNN) models for image classification. Additionally, the aim is to review the current state-of-the-art and to validate the system's functionality on two different datasets. A genetic algorithm is used to find the best approximate DNN model. Additionally, several approaches to encode the genetic information of DNN models are explored. Furthermore, several experiments with the VGG-16 architecture were conducted to find the best possible system base. The thesis also includes a discussion on the practice of model training and how problems that can arise during the automatic training of DNN models are avoided. The implementation is written in Python with Tensorflow library.
|
|
|
Neural networks used in autonomous vehicles
Ryšavý, Jan ; Píštěk, Václav (referee) ; Kučera, Pavel (advisor)
This bachelor thesis deals with the use of neural networks in autonomous vehicles. The first part of the thesis presents the basic principles of neural networks and learning methods that are used in autonomous vehicles. Then the thesis describes the architecture and functions of neural networks. The second part of the thesis also describes the different types of autonomous vehicles, their classifications and an overview of the sensors used by autonomous vehicles. The last part of the thesis deals with the implementation of neural networks in ECUs using programming languages and libraries, and applications such as object detection and marker recognition.
|
guest ::
