|
|
Current Trends in Data Analytics and their Successful Enterprise Applications
RUBÁŠOVÁ, Anna
This bachelor thesis focuses on current trends in data analytics and their application to business practice. The thesis describes the development of data analytics and machine learning methods that are used to mine knowledge from data. A specific example demonstrating the use of machine learning deals with predicting the outcome of a binary classification problem, where historical data is used to predict future trends. The selected methods are implemented using a "no-code" web-based tool. A comparative analysis of the implemented methods is performed in the thesis. The methods are evaluated on the basis of their accuracy and computational efficiency, and then the most appropriate one is identified. Finally, the selected method is further optimized using a threshold value to achieve the best results.
|
|
|
Wavelet Filtering of ECG Signals
Staša, Josef ; Kolář, Radim (referee) ; Kozumplík, Jiří (advisor)
The aim of this project is introduction to problems about filtering of signals using wavelet transform. This method is very effective and in present it is often exploited. The main imposition is to use wavelet transform for filtering of ECG signals. Practical part of my project is realized wiener filtering based on DTWT in MATLAB software, to show advantages, disadvantages and compare linear filtering .
|
|
|
Analysis of EMG Signal for Prosthetic hand based on Fuzzy Logic Technique
Ollé, Tamás ; Kolářová, Jana (referee) ; Pathak, Pawan Kumar (advisor)
Lidská ruka je koncový orgán horní končetiny, který slouží k důležité funkci uchopení, stejně jako důležitý orgán pro vnímání a komunikaci. Je to úžasný příklad o tom, jak komplexní mechanismus může být implementovaný, schopný chápat velice komplexní a užitečné úkoly používáním velmi efektivní kombinace mechanismů, snímání a řídicích funkcí. Elektromyogram (EMG) byl původně vyvinutý pro vyšetrování svalnatého nepořádku. Klinické aplikace se brzo rozšířila, nejpozoruhodněji ve výzkumu epilepsii a nakonec se stal populární po zavedení protéz, specificky tělem poháněných protéz. EMG nahrávání je použito pro studování funkčního stavu svalu během různých pohybů. Cílem tohoto projektu je vyvinutí elektromyogramové (EMG) metody třídění, která bude pomáhat v aplikacích jako systém pracující v reálném čase. První část tohoto projektu bylo získávání informací. Údaje reálného času byly zaznamenaný pomocí EMG monitorovacího systému (BIOPAC) a kompletní datová sada různých osob byl zaznamena. Tyto EMG data byly konvertované ze souboru ASCII do čitelné formy pro MATLAB. Druhou částí projektu byla extrakce vlastností. Pět tradičních parametrických rysů bylo vypočítáno, jmenovitě Integrovaný EMG (IEMG), variance (VAR), Zero Crossing (ZC), Slope Sign Changes (SSC) a Vlnová délka (WL). Třetí část projektu byla klasifikace EMG vzorů s použitím fuzzy logiky. Výsledky jsou docela slibné.
|
|
|
Motion Detection in Video
Polanský, Petr ; Sochor, Jakub (referee) ; Herout, Adam (advisor)
The objective of this work is evaluate motion detection using Gaussian Mixture Model. When algorithm detects motion, it creates short video capturing this motion and visualizing it properly. Visualization is made by white pixels intensity graph. System is applicable on less busy area when motion is more noticeable. Results of this work shows how surrounding environment and camera position influence detection.
|
|
|
QRS detection using zero crossing counts
Kašák, Pavel ; Hrubeš, Jan (referee) ; Vítek, Martin (advisor)
This thesis is focusing on problems of detection of QRS complex. It includes basic description of ECG signal. It acquaints with common methods of searching QRS complex. Main attention is paid to detection based on zero crossing counts, which is realized by MATLAB program. This algorithm is then tested on CSE database, optimised and its efficiency is evaluated.
|
|
|
Detection of Traffic Signs and Lights
Oškera, Jan ; Špaňhel, Jakub (referee) ; Herout, Adam (advisor)
The thesis focuses on modern methods of traffic sign detection and traffic lights detection directly in traffic and with use of back analysis. The main subject is convolutional neural networks (CNN). The solution is using convolutional neural networks of YOLO type. The main goal of this thesis is to achieve the greatest possible optimization of speed and accuracy of models. Examines suitable datasets. A number of datasets are used for training and testing. These are composed of real and synthetic data sets. For training and testing, the data were preprocessed using the Yolo mark tool. The training of the model was carried out at a computer center belonging to the virtual organization MetaCentrum VO. Due to the quantifiable evaluation of the detector quality, a program was created statistically and graphically showing its success with use of ROC curve and evaluation protocol COCO. In this thesis I created a model that achieved a success average rate of up to 81 %. The thesis shows the best choice of threshold across versions, sizes and IoU. Extension for mobile phones in TensorFlow Lite and Flutter have also been created.
|
|
|
Universal Audio 1176 Compressor
Pavlik, Miroslav ; Ištvánek, Matěj (referee) ; Koton, Jaroslav (advisor)
The aim of this bachelor thesis is to get acquainted with the issue of analog compressor Universal Audio 1176 and verify the functionality and principles of individual block diagrams of this compressor. The bachelor's thesis is carried out for a detailed analysis of individual blocks of compressor 1176. A practical implementation was constructed, to verify the correct functionality of compressor 1176 on measurements, which were performed in the laboratory.
|
|
|
ECG signal filtering in the wavelet domain
Zedníček, Vlastimil ; Vítek, Martin (referee) ; Smital, Lukáš (advisor)
The aim of this work is introduction with the method of filtering the signal using wavelet transform. Our task was to use a redundant wavelet transform for filtering ECG signals. First we met with wavelet filtering. In the practical part, we implemented a filter based on redundant DTWT in Matlab. The results were evaluated based on the achieved signal / noise ratio. The proposed filter, we tested the CSE database, and we tried to optimize the filter parameters and the degree of decomposition of wavelet transform, the bank decomposition and reconstruction filters, thresholding methods and threshold setting. The results were evaluated based on the achieved signal / noise ratio. as a last resort, we compared the results obtained with the results of different filtering methods
|
|
| |
|
|
Explainable Convolutional Neural Networks
Kamenický, Daniel ; Herout, Adam (referee) ; Hradiš, Michal (advisor)
The aim of this work was to compare several methods for visualizing the features of each class on the input pixel layer of the CNN. Each method uses a different algorithm, based on gradients, to compute the resulting values. Using the implementation of each method, the resultant values of the methods are obtained by using the equation of energy concentration. The resultant values are presented in tables and graphs from which the success rate of the result of the work can be read. The difference between the methods and comparison of their results can be read from the work. This makes it possible to get an overview of gradient based visualization methods.
|
guest ::
