Humboldt-Univ., Diss.--Berlin, 2003. / Literaturverz. S. [205] - 212.

The aim of this study is to get some approach which can help in improving the predictions of daily temperature of Umeå. Weather forecasts are available through various sources nowadays. There are various software and methods available for time series forecasting. Our aim is to investigate the daily maximum temperatures of Umeå, and compare the performance of some methods in forecasting these temperatures. Here we analyse the data of daily maximum temperatures and find the predictions for some local period using methods of autoregressive integrated moving average (ARIMA), exponential smoothing (ETS), and cubic splines.  The forecast package in R is used for this purpose and automatic forecasting methods available in the package are applied for modelling with ARIMA, ETS, and cubic splines. The thesis begins with some initial modelling on univariate time series of daily maximum temperatures. The data of daily maximum temperatures of Umeå from 2008 to 2013 are used to compare the methods using various lengths of training period. On the basis of accuracy measures we try to choose the best method. Keeping in mind the fact that there are various factors which can cause the variability in daily temperature, we try to improve the forecasts in the next part of thesis by using multivariate time series forecasting method on the time series of maximum temperatures together with some other variables. Vector auto regressive (VAR) model from the vars package in R is used to analyse the multivariate time series. Results: ARIMA is selected as the best method in comparison with ETS and cubic smoothing splines to forecast one-step-ahead daily maximum temperature of Umeå, with the training period of one year. It is observed that ARIMA also provides better forecasts of daily temperatures for the next two or three days. On the basis of this study, VAR (for multivariate time series) does not help to improve the forecasts significantly. The proposed ARIMA with one year training period is compatible with the forecasts of daily maximum temperature of Umeå obtained from Swedish Meteorological and Hydrological Institute (SMHI).

This thesis contains theory of probability and statistical sets. Solved and unsolved problems of probability, random variable and distributions random variable, random vector, statistical sets, regression and correlation analysis. Unsolved problems contains solutions.

Tato diplomová práce se zabývá problematikou vzdáleného rozpoznávání mluvčích. V případě dat zachycených odlehlým mikrofonem se přesnost standardního rozpoznávání značně snižuje, proto jsem navrhl dva přístupy pro zlepšení výsledků. Prvním z nich je použití mikrofonního pole (záměrně rozestavené sady mikrofonů), které je schopné nasměrovat virtuální "paprsek" na pozici řečníka. Dále jsem prováděl adaptaci komponent systému (PLDA skórování a extraktoru i-vektorů). S využitím simulace pokojových podmínek jsem syntetizoval trénovací a testovací data ze standardní datové sady NIST 2010. Ukázal jsem, že obě techniky a jejich kombinace vedou k výraznému zlepšení výsledků. Dále jsem se zabýval společným určením identity a pozice mluvčího. Zatímco výsledky ve venkovním simulovaném prostředí (bez ozvěn) jsou slibné, výsledky z interiéru (s ozvěnami) jsou smíšené a vyžadují další prozkoumání. Na závěr jsem mohl systémem vyhodnotit omezené množství reálných dat získaných přehráním a záznamem nahrávek ve skutečné místnosti. Zatímco výsledky pro mužské nahrávky odpovídají simulaci, výsledky pro ženské nahrávky nejsou přesvědčivé a vyžadují další analýzu.

In this master thesis is described how to design and implement classifier of defects detected during the final stage of production nonwovens. The beginning of the thesis is devoted to the analysis of options for image processing and classification. Followed by the part, where is described process of image segmentation and extraction of feature vector. Description of classifier implementation and table of achieved results of classification on real images of detected defects.

The aim of this work is an optimalization of vector contact force sensor properties which prototype is at the moment designed. The principle of sensor is based on transformation active part of sensor using three tensiometers and following value indentification of loading force vector using neuron network. Sensor has a bad sensitivity when axis force load or near to axis of sensor. To solve this problem ''MKP'' model of sensor was used on which suitable optimalization method was applicated. From the result of optimalization modificated sensor will be proposed and will be made a utility verification using neuron network.

This dissertation presents the solution of difference equations and focuses on a method of difference equations solution with the aid of eigenvectors. The first part reminds the basic terms from area of difference equations such as dynamic of difference equations and linear difference equations of first order and higher order. Then the second section recalls also the system of difference equations including the fundamental matrix and general solution description. Afterthat, the method of solving the difference equations with a variation of constants and transform of scalar equations to the system are shown. The second part of the dissertation analyses some known algorithms and methods for the solution of linear difference equations. The Z-transform, its importance and usage for finding the solution of difference equation is recalled. Then the discrete analogue of Putzer's algorithm is mentioned because this algorithm was often used to check the results obtained by the newly described algorithm in further parts of this thesis. Also some ways of the system matrix power are stated. The next section then describes the principle of Weyr's method which is the basic point for further development of the theory including the presentation of the research results gained by Jiří Čermák in this area. The third part describes own solution of the difference equations system via eigenvectors based on the principle of Weyr's method for differential equations. The solution of system of linear homogeneous difference equtions with constant coefficients including the proof is presented and this solution is then extended to nonhomogeneous systems. Consequently to the theory, the influence of a nulity and the multiplicity of roots on the form of the solution is discussed. The last section of this part shows the implementation of the algorithm in Matlab program (for basic simpler cases) and its application to some cases of difference equations and systems with these equations. The final part of the thesis is more practical and it presents the usage of the designed algorithm and theory. Firstly, the algorithm is compared with Z-transform and the method of variation of constants and it is illustrated how to obtain the same results by using these three approaches. Then an example of current response solution in RLC circuit is demonstrated. The continuous case is solved and then the problem is transferred to discrete case and solved with the Z-transform and the method of eigenvectors. The obtained results are compared with the result of the continuous case.

Tkáňová optika nabývá rychle na významu a přesná znalost optických vlastností biologických tkání je podstatná pro výzkum v biomedicíně i pro kontrolu kvality potravin. Jestliže je vzorek tkáně osvětlen, dochází k mnohonásobnému odrazu světla. V případě postmortem neživých tkání (maso) je rozměr buněk větší než vlnová délka použitého světla. Dochází k Mieovu rozptylu prošlého nebo zpět odraženého světla, v důsledku čehož se objevují různé polarizační stavy světla. Polarizační stavy světla rozptýleného na difúzním prostředí jsou experimentálně zkoumány a modelovány. V práci jsme provedli dva modifikované experimenty: rozptyl polarizovaného světla, které dvakrát prochází vzorkem (vpřed a vzad) a jen světla, které jen prochází vzorkem. Měření rozptýleného světla ukazuje, že dochází k depolarizaci a ke stáčení polarizační roviny, což obojí závisí na orientaci svalových vláken a stárnutí tkání postmortem. Mimo experimentů byl také proveden teoretický popis difúzní biologické tkáně a byla vypočtena radiační přenosová rovnice pomocí modifikované Monte Carlo metody, která zahrnuje polarizační stav světla (PLMC). Je ukázáno, že stupeň polarizace podstatně závisí na optických vlastnostech rozptylového prostředí. Výsledky ukazují, že stav polarizace světla na výstupu závisí na stavu polarizace světla před vzorkem a na optických vlastnostech a tloušťce vrstvy prostředí v průběhu jejího stárnutí. Je také provedena korelace změn polarizace na čerstvosti masa, i popis dynamického chování polarizace při stárnutí masa.

This master's thesis deals with the geographic information systems   with an emphasis on so - called vector analysis . The aim of the thesis is to create a library that implements the internal representation   of geographic data and performs analytical operation over them .  The thesis describes the logical structure of the application and the hierarchy of abstract classes . In the form of pseudo -code is shown a set of individual computing operations . Work is strongly based on geographic system GRASS. This GIS is connectable through  import and export vector data in shapefile format. Concept of   storing layers is designed for ease of data manipulation and  their portability . The main idea behind the application is to split the library into individual modules . These modules are  independent , but also can benefit from each other's functions . In conclusion are demonstrated implemented vector analysis and  representation of their calculation .

In this paper, we propose techniques for adaptation of speaker recognition systems. The aim of this work is to create adaptation for Probabilistic Linear Discriminant Analysis. Special attention is given to unsupervised adaptation. Our test shows appropriate clustering techniques for speaker estimation of the identity and estimation of the number of speakers in adaptation dataset. For the test, we are using NIST and Switchboard corpora.