In the scheme of Non Destructive Testing (NDT), defect detection is an important process. Traditional image processing techniques have successfully been used for defect recognition. Usage of machine learning techniques is still in the initial stages of development. Convolution Neural Networks (CNN) is widely used for object classification one such scenario is defect classification in weld tubes. With the advent of deep learning techniques such as transfer learning, we can transfer knowledge gained in one domain successfully into other. Pre-trained models successfully learn features from large scale datasets that can be used for in domains having sparse data and smaller datasets. The aim of this work is to help a manual inspector in recognition of defects on the weld tubes. With a given set of images, we proceed by forming unique pipeline architecture for automatic defect recognition. The research in this thesis focuses on extraction of welds using image segmentation techniques, creating a dataset of defects and using it to on pre-trained Convolution Neural Networks of VGG16, VGG19, Inception V3 and ResNet101. We evaluate the models on different metrics finding the best suited model for the created dataset. Further a prototype sliding window solution is used to find defects over the extracted weld region. We also present the limitations of this approach and suggest modifications that can be implemented in the future.

Memristor is a novel passive electronic device and a promising candidate for new generation non-volatile memory and analog computing. Analog computing based on memristors has been explored in this study. Due to the lack of commercial electrical testing instruments for those emerging devices and crossbar arrays, we have designed and built testing circuits to implement analog and parallel computing operations. With the setup developed in this study, we have successfully demonstrated image processing functions utilizing large memristor crossbar arrays. We further designed and experimentally demonstrated the first memristor based field programmable analog array (FPAA), which was successfully configured for audio equalizer and frequency classifier demonstration as exemplary applications of such memristive FPAA (memFPAA).

In this thesis there is a design of universal time-variant convolution FIR processor. Proposed algorithm is ready to process long signals with reduced process delay. Furthermore, there are proposed methods for smooth crossfade between different impulse responses. Algorithm is implemented in Matlab environment and later as a VST plugin.

This diploma thesis deals with the application of deep neural networks with focus on image segmentation. The theoretical part contains a description of deep neural networks and a summary of widely used convolutional architectures for segmentation of objects from the image. Practical part of the work was devoted to testing of an existing network architectures. For this purpose, an open-source software library Tensorflow, implemented in Python programming language, was used. A frequent problem incorporating the use of convolutional neural networks is the requirement on large amount of input data. In order to overcome this obstacle a new data set, consisting of a combination of five freely available databases was created. The selected U-net network architecture was tested by first modification of the newly created data set. Based on the test results, the chosen network architecture has been modified. By these means a new network has been created achieving better performance in comparison to the original network. The modified architecture is then trained on a newly created data set, that contains images of different types taken with various fundus cameras. As a result, the trained network is more robust and allows segmentation of retina blood vessels from images with different parameters. The modified architecture was tested on the STARE, CHASE, and HRF databases. Results were compared with published segmentation methods from literature, which are based on convolutional neural networks, as well as classical segmentation methods. The created network shows a high success rate of retina blood vessels segmentation comparable to state-of-the-art methods.

This diploma thesis focuses on a face recognition from a video, specifically how to aggregate feature vectors into a single discriminatory vector also called a template. It examines the issue of the extremely angled faces with respect to the accuracy of the verification. Also compares the relationship between templates made from vectors extracted from video frames and vectors from photos. Suggested hypothesis is tested by two deep convolutional neural networks, namely the well-known VGG-16 network model and a model called Fingera provided by company Innovatrics. Several experiments were carried out in the course of the work and the results of which confirm the success of proposed technique. As an accuracy metric was chosen the ROC curve. For work with neural networks was used framework Caffe.

The actual traffic flow theory assumes interactions only between neighbouring vehicles within the traffic. This assumption is reasonable, but it is based on the possibilities of science and technology available decades ago, which are currently overcome. Obviously, in general, there is an interaction between vehicles at greater distances (or between multiple vehicles), but at the time, no procedure has been put forward to quantify the distance of this interaction. This work introdukce a method, which use mathematical statistics and precise measurement of time distances of individual vehicles, which allows to determine these interacting distances (between several vehicles) and its validation for narrow densities of traffic flow. It has been revealed that at high traffic flow densities there is an interaction between at least three consecutive vehicles and four and five vehicles at lower densities. Results could be applied in the development of new traffic flow models and its verification.

This master´s thesis deals with blood vessel segmentation in retinal image data. The theoretical part is focused on the basic description of anatomy and physiology of the eye and methods of observing the back of the eye. This thesis also describes the principles of classical and convolutional neural networks and segmentation techniques that are used to segment blood vessel in retinal images. In the practical part, a segmentation method using convolutional neural network U-net is implemented. This neural network is trained on the three datasets. Two datasets include images from experimental video ophthalmoscope. Because it impossible to compare the results of these two datasets with any other methods of retinal blood vessel segmentation, U-net is trained on other dataset that is HRF database. This dataset includes fundus images. The results of testing on this dataset serves for comparing results with other methods of retinal blood vessel segmentation.

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.

The aim of this work is to increase the accuracy of the transcription of text documents. This work is mainly focused on texts printed on degraded materials such as newspapers or old books. To solve this problem, the current method and problems associated with text recognition are analyzed. Based on the acquired knowledge, the implemented method based on GAN network architecture is chosen. Experiments are a performer on these networks in order to find their appropriate size and their learning parameters. Subsequently, testing is performed to compare different learning methods and compare their results. Both training and testing is a performer on an artificial data set. Using implemented trained networks increases the transcription accuracy from 65.61 % for the raw damaged text lines to 93.23 % for lines processed by this network.

This master’s thesis deals with designs and implementation of systems for music cover recognition. The introduction part is devoted to the calculation parameters from audio signal using Music Information Retrieval techniques. Subsequently, various forms of cover versions and musical aspects that cover versions share are defined. The thesis also deals in detail with the creation and distribution of a database of cover versions. Furthermore, the work presents methods and techniques for comparing and processing the calculated parameters. Attention is then paid to the OTI method, CSM calculation and methods dealing with parameter selection. The next part of the thesis is devoted to the design of systems for recognizing cover versions. Then there are compared systems already designed for recognizing cover versions. Furthermore, the thesis describes machine learning techniques and evaluation methods for evaluating the classification with a special emphasis on artificial neural networks. The last part of the thesis deals with the implementation of two systems in MATLAB and Python. These systems are then tested on the created database of cover versions.