|
|
Adaptive Sampling of Input Packets Implemented in FlowMon Probe
Kaštovský, Petr ; Martínek, Tomáš (referee) ; Kořenek, Jan (advisor)
There is a FlowMon probe being developed in a Libeouter project that is used for passive network measurements. The probe has better stability and accuracy than sofware based solutions even under a heavy load or network attack. To guarantee a precision of results there is a need to data reduction to prevent measuring system overload. There are few kinds of data reduction. Method used in the FlowMon probe is called sampling. Adaptive sampling unit sets the sampling rate (rate of processed and discarded packets) according to actual state of measured network.
|
|
|
QRS Complex Detector
Kubík, Adam ; Provazník, Ivo (referee) ; Smital, Lukáš (advisor)
The aim of my thesis is to introduce issue of detection of QRS complex by squaring and envelope detection. The first part focuses on creation and scanning of heart electrical activity. Afterwards, the most frequent types of disturbance of EKG signal and their filtering are described. The second part is about real application of signal filtration and emphasis of QRS complexes using the methods which are mentioned above. Practical part is externalized in Matlab program. And signals from CSE library are used to test efficiency of the detectors.
|
|
|
Postprocessing of GPS Logs
Jaška, Roman ; Pavelková, Alena (referee) ; Polok, Lukáš (advisor)
This thesis aims to analyse common errors in recorded GPS activities, identify algorithms suitable for correction of these problems and design and subsequently implement an Android application. The first chapter contains an introduction to the problem, demonstration of a few affected examples and explains the motivation behind the choice of this subject for my thesis. Second chapter decomposes the established problem. I specifically elaborate on the source of the errors, the format of input data, existing solutions, mathematical methods suitable for this problem, available APIs and reasons behind the selection of the final platform. Chapter number three describes the actual solution of the problem employed in the final application. The requirements for the final application are laid out as well as the specific details of the functionality, such as the usage of Kalman filter, individual APIs, libraries as well as the design of the user interface. The penultimate chapter reveals selected parts of the implementation, such as the Kalman filtering itself or the problem of manual track correction. This chapter also contains description of design changes executed during the implementation. The final chapter reviews the achieved results of Kalman filtering and snapping of points to roads using Google Maps Roads API. Ultimately, I elaborate on possibilities of continued development of the application.
|
|
|
Data Analysis and Clasification from the Brain Activity Detector
Ullrich, Petr ; Šůstek, Martin (referee) ; Szőke, Igor (advisor)
This thesis deals with the issue of recording brain activity, implementation of its processing, analysis and classification. The OpenBCI hardware is used for recording. I have studied and described necessary information about recording brain activity and OpenBCI project. Design for data set, data processing and thoughts classification was created. Created system allows classification based on recorded brain activity. The neural network was used for classfication, but the success of the recognition of designed classes was not high.
|
|
|
Open-source systems for image processing
Slinták, Vlastimil ; Zukal, Martin (referee) ; Mikulka, Jan (advisor)
The aim of this document is to describe open-source libraries for image processing — ITK, VTK and OpenCV. Their architectures are described and it is shown how to compile and use them in real applications. It also provides basic description of algorithms for image processing. In the second part of this document, we try to use ITK library with data-mining environment RapidMiner.
|
|
| |
|
|
Measurement of conifers needles surface by MR tomography
Hladík, Roman ; Kubásek, Radek (referee) ; Mikulka, Jan (advisor)
This bachelor thesis deals with the tomography based on magnetic resonance and evaluation of scanned images of conifers needles. At first, it is described the physical basis and methods of magnetic resonance scanning of objects. Another part is devoted to graphical preprocessing, which consist of noise suppression and image segmentation. After that there are described the design and algorithm for parameter estimation implemented to the application. The first experimental measurements were lightly discussed and evaluated in the last part of the work. In the last part there are discussed results of the experimental measurement. The attached documents contain the source code in Matlab language.
|
|
|
Passive shunt filtration and compensation
Urant, Tomáš ; Novotný, Jan (referee) ; Drápela, Jiří (advisor)
The aim of this work is to acquaint the reader with the theoretical and practical design filtration-compensation device. The theoretical part describes the problems of reactive power compensation and harmonic filtering components. There are described various methods and implementation of compensation and filtration-compensation devices. In the practical part is the theoretical basis used for the realization of a laboratory model of filtration-compensation device for precisely specified load with precisely specified requirements. In laboratory model can be seen meaning of filtering harmonics by using of simple hand switching individual filters.
|
|
|
IP Flow Filtration and Profiling
Sedlák, Michal ; Tisovčík, Peter (referee) ; Kučera, Jan (advisor)
This thesis addresses the problem of filtering and profiling IP flows, primarily data of IPFIXsystems. Within the work, a general filtering component is designed and implemented, whichaims to be sufficiently efficient and flexible for use in other projects related to IP flows. Thiscomponent is then adapted to work with data in the IPFIX protocol format and integratedinto the existing modular collector IPFIXcol2 in the form of plugins adding the support forfiltering of passing IPFIX data and their sorting into profiles.
|
|
|
Information Retrieval in Czech Wikipedia
Balgar, Marek ; Bartík, Vladimír (referee) ; Chmelař, Petr (advisor)
The main task of this Masters Thesis is to understand questions of information retrieval and text classifi cation. The main research is focused on the text data, the semantic dictionaries and especially the knowledges inferred from the Wikipedia. In this thesis is also described implementation of the querying system, which is based on achieved knowledges. Finally properties and possible improvements of the system are talked over.
|
guest ::
