|
|
Image watermarking
Štágl, Luboš ; Zezula, Radek (referee) ; Číka, Petr (advisor)
This diploma thesis is concerned in static pictures Security problems. That means, additional informations (watermark) are embedded to the original picture in a specific way. This complex picture structure (watermark) should be unremoval by different attacks runing processes. The main goal of this diploma thesis is realize two separated methods digital data watermarking in MATLAB program. For reason of a quite large scale of different watermarking methods are chosen at this time only two of take were choosen. First of the methods is Watermark injection in the spatial domain and the second in the freuquency domain. Both methods are set up in a special way and finish goals of the process. Expected result are, that digital picture user doesn't know about the watermarking technique was aplied on this picture and the watermarking data are the most resistant as possible as can be. These attacks were simulated in Checkmark program.
|
|
|
Super-resolution methods
Franěk, Pavel ; Fedra, Petr (referee) ; Mézl, Martin (advisor)
The main goal of this bachelor’s thesis is acquaint with method, which enable increasing resolution digital photos. Also realize individual interpolation method and Super-resolution by the help of programme Matlab and reference on estimation record. Discuss possibility using method super- resolution for imagery with medical modality.
|
|
|
Robust portfolio selection
Horváthová, Inés ; Červinka, Michal (advisor) ; Kraicová, Lucie (referee)
In this thesis, we take the mean-risk approach to portfolio optimi- zation. We will first define risk measures in general and then intro- duce three commonly used ones: variance, Value-at-risk (V aR) and Conditional-value-at-risk (CV aR). For each of these risk measures we formulate the corresponding mean-risk models. We then present their robust counterparts. We focus mainly on the robust mean-variance models, which we also apply to historical data using free statistical software R. Finally, we compare the results with the classical non- robust mean-variance model.
|
|
|
Super-resolution methods
Franěk, Pavel ; Fedra, Petr (referee) ; Mézl, Martin (advisor)
The main goal of this bachelor’s thesis is acquaint with method, which enable increasing resolution digital photos. Also realize individual interpolation method and Super-resolution by the help of programme Matlab and reference on estimation record. Discuss possibility using method super- resolution for imagery with medical modality.
|
|
|
Image watermarking
Štágl, Luboš ; Zezula, Radek (referee) ; Číka, Petr (advisor)
This diploma thesis is concerned in static pictures Security problems. That means, additional informations (watermark) are embedded to the original picture in a specific way. This complex picture structure (watermark) should be unremoval by different attacks runing processes. The main goal of this diploma thesis is realize two separated methods digital data watermarking in MATLAB program. For reason of a quite large scale of different watermarking methods are chosen at this time only two of take were choosen. First of the methods is Watermark injection in the spatial domain and the second in the freuquency domain. Both methods are set up in a special way and finish goals of the process. Expected result are, that digital picture user doesn't know about the watermarking technique was aplied on this picture and the watermarking data are the most resistant as possible as can be. These attacks were simulated in Checkmark program.
|
|
|
Robust portfolio optimization model
Löw, Alexandr ; Pelikán, Jan (advisor) ; Fábry, Jan (referee)
Portfolio optimization models aim to optimally distribute capital among selected stocks, bonds and other securities and financial products offered on financial markets. An important factor in the optimization is the risk, which is by definition very abstract concept and is very difficult to quantify. Robust portfolio optimization model is based on the general robust binary model. The robustness of this model lies in a two-stage optimization, where every solution is subject to maximization of losses and from these pessimistic estimates such a solution is selected that best meets the user's criteria, in our case total return of the portfolio.
|
|
|
Bayesian vector auto-regression model with Laplace errors applied to financial market data
Šindelář, Jan
The article presents alternative version of Bayesian vector auto-regression model with Laplace distributed innovations. Bayesian estimation in such model is more computationally demanding than estimation in a model with normally distributed innovations, but because of the heavier tails of Laplace distribution, it is more robust. In the article I try to present the way of proceeding with the estimation, obtaining a full posterior distribution of the parameters as a result. At the end an efficient algorithm is sketched, but this part of the research is still unfinished.
|
guest ::
