|
|
Image based smoke and fire detection
Ďuriš, Denis ; Burda, Karel (referee) ; Přinosil, Jiří (advisor)
This diploma thesis deals with the detection of fire and smoke from the image signal. The approach of this work uses a combination of convolutional and recurrent neural network. Machine learning models created in this work contain inception modules and blocks of long short-term memory. The research part describes selected models of machine learning used in solving the problem of fire detection in static and dynamic image data. As part of the solution, a data set containing videos and still images used to train the designed neural networks was created. The results of this approach are evaluated in conclusion.
|
|
|
Prediction of data flow in computer networks
Zvěřina, Lukáš ; Sobek, Jiří (referee) ; Vychodil, Petr (advisor)
The aim of this thesis was to study problems of prediction of data in computer networks. Furthermore, this work deals with network traffic and analyzing its properties. In this study were analyzed the possibilities of network traffic prediction using Farima model, the theory of chaos with Lyapunov exponents and neural networks. Possibilities of prediction with the focus on neural network were discussed in detail here, mainly on recurrent neural networks. Prediction was performed in Matlab development environment in Neural Network Toolbox, where they were created, trained and evaluated neural network to predict specific types of network traffic. For testing were selected Elman network NARX network and general LRN recurrent network. The results were clearly organized into tables and plotted in graphical relationships before and after the use of predictive techniques designed to final evaluation.
|
|
|
Neural Network for Autocomplete in the Browser
Kubík, Ján Jakub ; Zemčík, Pavel (referee) ; Kolář, Martin (advisor)
The goal of this thesis is to create and train a neural network and use it in a web browser for English text sequence prediction during writing of text by the user. The intention is to simplify the writing of frequent phrases. The problem is solved by employing a recurrent neural network that is able to predict output text based on the text input. Trained neural network is then used in a Google Chrome extension. By normalized ouput of the neural network, text choosing by sampling decoding algorithm and connecting, the extension is able to generate English word sequences, which are shown to the user as suggested text. The neural network is optimized by selecting the right loss function, and a suitable number of recurrent layers, neurons in the layers, and training epochs. The thesis contributes to enhancing the everyday user experience of writing on the Internet by using a neural network for English word sequence autocomplete in the browser.
|
|
|
Algorithmic Accompaniment Composition
Vinš, Jakub ; Hradiš, Michal (referee) ; Kolář, Martin (advisor)
This thesis deals with problems of computer music, especially with generating accompaniment to an existing song in MIDI format by means of artificial neural networks. Existing methods of algorithmic music composition are presented in the beginning. Followed by problems and their solutions connected with the conversion of MIDI files to matrices, which are suitable as an input for neural network and their inverse transformation. Subsequently are proposed, created, optimized and evaluated models which generate saxophone and piano accompaniment by means of feedforward and recurrent neural network. At the end model generates accompaniment to my own song as a form of a test.
|
|
|
Advanced classification of cardiac arrhythmias in ECG
Sláma, Štěpán ; Hejč, Jakub (referee) ; Ronzhina, Marina (advisor)
This work focuses on a theoretical explanation of heart rhythm disorders and the possibility of their automatic detection using deep learning networks. For the purposes of this work, a total of 6884 10-second ECG recordings with measured eight leads were used. Those recordings were divided into 5 groups according to heart rhythm into a group of records with atrial fibrillation, sinus rhythms, supraventricular rhythms, ventricular rhythms, and the last group consisted of the others records. Individual groups were unbalanced represented and more than 85 % of the total number of data are sinus rhythm group records. The used classification methods served effectively as a record detector of the largest group and the most effective of all was a procedure consisting of a 2D convolutional neural network into which data entered in the form of scalalograms (classification procedure number 3). It achieved results of precision of 91%, recall of 96% and F1-score values of 0.93. On the contrary, when classifying all groups at the same time, there were no such quality results for all groups. The most efficient procedure seems to be a variant composed of PCA on eight input signals with the gain of one output signal, which becomes the input of a 1D convolutional neural network (classification procedure number 5). This procedure achieved the following F1-score values: 1) group of records with atrial fibrillation 0.54, 2) group of sinus rhythms 0.91, 3) group of supraventricular rhythms 0.65, 4) group of ventricular rhythms 0.68, 5) others records 0.65.
|
|
|
Nonlinear analysis and prediction of network traffic
Člupek, Vlastimil ; Burget, Radim (referee) ; Vychodil, Petr (advisor)
This thesis deal with an analysis of network traffic and its properties. In this thesis are discussed possibilities of prediction network traffic by FARIMA model, theory of chaos with Lyapunov exponent and by neural networks. The biggest attention was dedicated to prediction network traffic by neural networks. In Matlab with using Neural Network Toolbox were created, trained and tested recurrent networks for prediction specific types of network traffics, which was captured on local network. There were choosed Elman network, LRN and NARX network to test the prediction of network traffic, results were discussed. Thesis also introduce area of application ability prediction of network traffic, there is introduce design of system for dynamic allocation bandwidth with particular description of its prediction part. Thesis also states possible use designed system for dynamic allocation of bandwidth.
|
|
|
Recurrent Neural Networks with Elastic Time Context in Language Modeling
Beneš, Karel ; Veselý, Karel (referee) ; Hannemann, Mirko (advisor)
Tato zpráva popisuje experimentální práci na statistické jazykovém modelování pomocí rekurentních neuronových sítí (RNN). Je zde předložen důkladný přehled dosud publikovaných prací, následovaný popisem algoritmů pro trénování příslušných modelů. Většina z popsaných technik byla implementována ve vlastním nástroji, založeném na knihovně Theano. Byla provedena rozsáhlá sada experimentů s modelem Jednoduché rekurentní sítě (SRN), která odhalila některé jejich dosud nepublikované vlastnosti. Při statické evaluaci modelu byly dosažené výsledky relativně cca. o 2.7 % horší, než nejlepší publikované výsledky. V případě dynamické evaluace však bylo dosaženo relativního zlepšení o 1 %. Dále bylo experimentováno i s modelem Strukturně omezené rekurentní sítě, ale ten se nepodařilo natrénovat k předpokládáným výkonům. Konečně bylo navrženo rozšíření SRN, pojmenované Náhodně prořidlá rekurentní neuronová síť. Experimentálně bylo potvrzeno, že RS-RNN dosahuje lepších výsledků v učení vlastního trénovacího korpusu a kombinace několika RS-RNN modelů přináší o 30 % větší zlepšení než kombinace stejného počtu SRN.
|
|
|
Convolutional Networks for Historic Text Recognition
Vešelíny, Peter ; Kolář, Martin (referee) ; Kišš, Martin (advisor)
This thesis deals with text line recognition of historical documents. Historical texts dating back to the 17th - 19th centuries are written in fraktur typeface. The character recognition problem is solved using neural network architecture called sequence-to-sequence . This architecture is based on encoder-decoder model and contains attention mechanism. In this thesis a dataset, from texts originated from German archiv called Deutsches Textarchiv , was created. This archive contains 3 897 different German books that have available transcripts and corresponding images of pages. The created dataset was used to train and experiment with the proposed neural network. During the experiments, several convolutional models, hyperparameters and the effects of positional embedding were investigated. The final tool can recognize characters with accuracy 99,63 %. The contribution of this work is the~mentioned dataset and neural network, which can be used to recognize historical documents.
|
|
|
Automatic tagging of musical compositions using machine learning methods
Semela, René ; Galáž, Zoltán (referee) ; Kiska, Tomáš (advisor)
One of the many challenges of machine learning are systems for automatic tagging of music, the complexity of this issue in particular. These systems can be practically used in the content analysis of music or the sorting of music libraries. This thesis deals with the design, training, testing, and evaluation of artificial neural network architectures for automatic tagging of music. In the beginning, attention is paid to the setting of the theoretical foundation of this field. In the practical part of this thesis, 8 architectures of neural networks are designed (4 fully convolutional and 4 convolutional recurrent). These architectures are then trained using the MagnaTagATune Dataset and mel spectrogram. After training, these architectures are tested and evaluated. The best results are achieved by the four-layer convolutional recurrent neural network (CRNN4) with the ROC-AUC = 0.9046 ± 0.0016. As the next step of the practical part of this thesis, a completely new Last.fm Dataset 2020 is created. This dataset uses Last.fm and Spotify API for data acquisition and contains 100 tags and 122877 tracks. The most successful architectures are then trained, tested, and evaluated on this new dataset. The best results on this dataset are achieved by the six-layer fully convolutional neural network (FCNN6) with the ROC-AUC = 0.8590 ± 0.0011. Finally, a simple application is introduced as a concluding point of this thesis. This application is designed for testing individual neural network architectures on a user-inserted audio file. Overall results of this thesis are similar to other papers on the same topic, but this thesis brings several new findings and innovations. In terms of innovations, a significant reduction in the complexity of individual neural network architectures is achieved while maintaining similar results.
|
|
|
Vehicle Classification Using Radar
Gottwald, Vilém ; Zemčík, Pavel (referee) ; Maršík, Lukáš (advisor)
Cílem této práce je rozpoznávání vozidel z radarových mračen bodů. Radar poskytuje informace o vzdálenosti a úhlu každého detekovaného cíle. Tyto informace lze převést do kartézského souřadnicového systému a získat tak 3D reprezentaci scény ve formě mračna bodů. V této práci jsou představeny stávající přístupy k rozpoznávání objektů v mračnech bodů. Metoda zvolená pro tuto práci spočívá v detekci objektů pomocí shlukování bodů a následné klasifikaci pomocí rekurentní neuronové sítě. Shluky bodů reprezentující objekty jsou vytvářeny z mračen bodů pomocí modifikovaného algoritmu DBSCAN. Z jednotlivých objektů jsou extrahovány příznaky, které jsou využity pro klasifikaci na různé typy vozidel pomocí neuronové sítě s dlouhodobou krátkodobou pamětí (LSTM). Pro trénování a vyhodnocení modelu byla vytvořena datová sada obsahující 57 345 anotovaných objektů. Vyvinutý model dosáhl na těchto datech 83% přesnosti metriky F1-skóre.
|
guest ::
