|
|
Autoencoder Implementation for Image Analysis
Sarančuk, Nikola ; Bilík, Šimon (referee) ; Horák, Karel (advisor)
The paper is devoted to the research of the problem of anomaly detection in industrial inspection. The paper describes the artificial neural network and its parts. The thesis contains a chapter where unary, binary and multi-class classifiers are compared. The thesis then explaines architecture of convolutional neural networks and autoencoder neural networks.. Then the paper describes the annotated dataset created. Finally, the paper describes the implementation of the convolutional autoencoder and evaluates the classification quality.
|
|
|
Advanced image analysis using deep neural networks
Hynek, Vojtěch ; Přinosil, Jiří (referee) ; Kiac, Martin (advisor)
This bachelor thesis deals with the problem of object detection in images using a convolutional neural network. The result of this work is a custom dataset, a neural network model YOLOv4 and a script used to process the resulting model data. The dataset contains 8080 images on which 14 objects are annotated. The neural network model was reduced in depth, which significantly increased the speed of the detection itself. The script processing the resulting data calculates the 3D and GPS coordinates of the detected object in space. The paper concludes by summarizing the results of the model and at the same time suggesting how the quality of the dataset could be improved.
|
|
|
Segmentation of intracardial ECG
Řehoř, Jan ; Ronzhina, Marina (referee) ; Novotná, Petra (advisor)
This master´s thesis deals with the segmentation of intracardiac ECG recordings and is divided into several parts. The first part is connected with a theoretical acquaintance with the issue, such as how the heart works, what is an intracardiac ECG and a convolutional neural network. Other parts of the work are already formed by the practical part, ie, signal annotation and model design automatically segmenting the intracaridal record. After the practical part, the evaluation of the results of the solution continues, comparison with the solution of third parties and with foreign studies dealing with a similar topic. The last part of the work is a discussion and conclusion, which summarizes the results of the work.
|
|
|
Detection of selected cardiac arrhythmias in ECG
Němečková, Karolína ; Ředina, Richard (referee) ; Ronzhina, Marina (advisor)
This thesis deals with classification of ECG records focusing on less classifiable arrhythmia (atrial flutter, atriventricular block I. and II. degree). In the theoretical part of the thesis deep learning used in classification of ECG records with a focus on the convolutional neural networks are described. The database of ECG records with a brief description of detected arrhythmias is further described. The practical part implements the proposed convolutional neural network in the Python environment. The evaluation of the arrhythmia detection quality was done using mainly the F1 score. The results were discussed at the end of the thesis.
|
|
|
Detection of poorly differentiated cardiac arrhythmias
Kantor, Marek ; Ronzhina, Marina (referee) ; Novotná, Petra (advisor)
This thesis focusses on the detection methods of atrial fibrilation, atrial flutter and sinus rhythm from ECG. Thesis also concentrate on the description of this arrhythmias and the learning algorithms used. In this thesis are implemented several classification approaches. For extraction of features is used convolution neural network and classification artifitial neural network. Selected 1D CNN method achived classification accuracy global F1 - score is 91 %. Moreover, the proposed CNN optimized with GA appears to be fast shallow network with better accuracy than the deep network. Created model are used for classification other type of arrhythmias too.
|
|
|
Proportional counter measurement system
Kolář, Ondřej ; Matěj, Zdeněk (referee) ; Kubíček, Michal (advisor)
This work focuses on development of a measurment system for proportional counters that are used to measure ionizing radiation. The system allows for data acquisition from proportional counters in form of time waveforms for further analysis. Core of the system is a Red Pitaya STEMlab 125-14 board which is able to record fast signals and transfer them to a computer for further processing. The first part of this work briefly describes theory of ionizing radiation and propotional counters. In the following part the measuring unit and its improvements are described. The main part of this work is focused on development of new software and description of individual software pieces. In the last part a real measurement is depicted, gathered data are analyzed and a machine learning method is proposed as a solution for proportional counters data analysis.
|
|
|
Train Identification System at Railway Switches And Crossings Using Advanced Machine Learning Methods
Krč, Rostislav ; Vorel,, Jan (referee) ; Plášek, Otto (referee) ; Podroužek, Jan (advisor)
This doctoral thesis elaborates possibilities of automatic train type identification in railway S&C using accelerometer data. Current state-of-the-art was considered, including requirements stated by research projects such as S-Code, In2Track or Turnout 4.0. Conducted experiments considered different architectures of artificial neural networks (ANN) and statistically evaluated multiple use case scenarios. The resulting accuracy reached up to 89.2% for convolutional neural network (CNN), which was selected as a suitable baseline architecture for further experiments. High generalization capability was observed as models trained on data from one location were able to classify locomotive types in the other location. Further experiments evaluated the effect of signal filtering and denoising. Evaluation of allocated memory and processing time for pre-trained models proved feasibility for in-situ application with regard to hardware restrictions. Due to a limited amount of available accelerometer data, distribution grid power demand data were utilized for further refinement of the proposed CNN architecture. Deep multi-layer architecture with regularization techniques such as dropout or batch normalization provides state-of-the-art performance for time series classification problems. Class activation mapping (CAM) allowed an explanation of decisions made by the neural network. Presented results proved that train type identification directly in the S&C is possible. The CNN was selected as optimal architecture for this task due to high classification accuracy, automatic filtration, and pattern recognition capabilities, allowing for the incorporation of the end-to-end learning strategy. Moreover, direct on-site application of pre-trained models is feasible with respect to limitations of in-situ hardware. This thesis contributes to understanding the train type identification problem and provides a solid theoretical background for future research.
|
|
|
Object clasification based on its topology change using image processing
Zbavitel, Tomáš ; Věchet, Stanislav (referee) ; Krejsa, Jiří (advisor)
The aim of the present work is to select a suitable object classification method for the recognition of one-handed finger alphabet characters. For this purpose, a sufficiently robust dataset has been created and is included in this work. The creation of the dataset is necessary for training the convolutional neural network. Further more, a suitable topology for data classification was found. The whole work is implemented using Python and the open-source library Keras was used.
|
|
|
Classification of UAV hyperspectral images using deep learning methods
Řádová, Martina ; Potůčková, Markéta (advisor) ; Kupková, Lucie (referee)
Diploma thesis "Classification of UAV hyperspectral images using deep learning methods" focuses on the classification methods, namely convolutional neural networks (CNN), of hyperspectral (HS) images. Based on a thorough literature review, a comprehensive overview on CNN utilisation in remote sensing is assembled as a basis for identifying suitable methods for the specific task of this thesis. Two methods with an open solution in programming language Python were selected - Capsule Network and U-Net. The main aim of this work is to verify the suitability of chosen methods for the classification of hyperspestral images. The images were acquired by sensors with high spatial resolution carried by a UAV over Krkonoše Mts. tundra. Important step was to prepare input HS data (54 bands, 9cm) to have suitable form for entering the network. Not all the required results were achieved due to the complexity of the Capsule Network architecture. The U-Net method was used in purpose of comparing and verifying the results. Accuracies retrieved from the U-Net overcome results achieved by traditionally used machine learning methods (SVM, ML, RF, etc). Overall accuracy for U-Net was higher than 90% where other mentioned methods did not get over 88%. Especially classes block fields and dwarf pine achieved higher...
|
|
|
Anticurtaining - Image Filter for Electron Microscopy
Dvořák, Martin ; Dobeš, Petr (referee) ; Zemčík, Pavel (advisor)
Tomographic analysis produces 3D images of examined material in nanoscale by focus ion beam (FIB). This thesis presents new approach to elimination of the curtain effect by machine learning method. Convolution neuron network is proposed for elimination of damaged imagine by the supervised learning technique. Designed network deals with features of damaged image, which are caused by wavelet transformation. The outcome is visually clear image. This thesis also designs creation of synthetic data set for training the neuron network which are created by simulating physical process of the creation of the real image. The simulation is made of creation of examined material by milling which is done by FIB and by process displaying of the surface by electron microscope (SEM). This newly created approach works precisely with real images. The qualitative evaluation of results is done by amateurs and experts of this problematic. It is done by anonymously comparing this solution to another method of eliminating curtaining effect. Solution presents new and promising approach to elimination of curtaining effect and contributes to a better procedure of dealing with images which are created during material analysis.
|
guest ::
