Thesis deals with polarization multiplexing and how birefringence influences polarization multiplexing. It also deals with polarization of light and how it can be described by Stokes and Jones vectors and also with visualiating polarization on Poincaré sphere. It also describes how information is transmitted in optical systems. At the end of the thesis is described design for polarization multiplexing at the lab.

Optomechanical components are widely used in many optical experiments. This diploma thesis deals with design and manufacturing optomechanical components using 3D print technology. These are cheaper alternatives of commercial devices. In addition to 3D printed parts, minimum number of other components are used to assemble functional devices. Using simple experimental setups, the manufactured components are evaluated and compared with commercially available ones. The results show that they can be used in applications where high accuracy is not required. The second part is devoted to the design and manufacturing of a polarimeter as a mechanism combining electrical and mechanical components with 3D printed parts. The polarimeter is used to measure some properties of polarized light.

Tato diplomová práce pojednává o dvou algoritmech pro dolování z proudu dat - Very Fast Decision Tree (VFDT) a Concept-adapting Very Fast Decision Tree (CVFDT). Je vysvětlen princip klasifikace rozhodovacím stromem. Je popsána základní myšlenka konstrukce stromu Hoeffding Tree, který je základem pro algoritmy VFDT a CVFDT. Tyto algoritmy jsou poté rozebrány detailněji. Dále se tato práce zabývá návrhem algoritmu Genetického Programování (GP), který je použit pro vytváření klasifikátoru obrazových dat. Vytvořený klasifikátor je použit jako alternativní způsob klasifikace objektů v obraze ve frameworku Viola-Jones. V práci je rozebrána implementace algoritmů, které jsou implementovány v jazyce Java. Algoritmus GP je integrován do knihovny “Image Processing Extension” programu RapidMiner. Algoritmy VFDT a CVFDT jsou testovány na syntetických a reálných textových datech. Algoritmus GP je testován na klasifikaci obrazových dat a následně vytvořený klasifikátor je otestován na detekci obličejů v obraze.

This thesis deals with the image object detection using Haar--like features and AdaBoost algorithm. The text describes methods how to train and test an object detector. The main contributon of this thesis consists in creation image object detector in Java programming language. Created algorithms were integrated as plugin into the RapidMiner tool, which is widely used and known worldwide as tool for data mining. The thesis contains the instructions for created operators and few exaples for executing in RapidMiner tool. The functionality of image object detector was demonstrated on selected medical images.

With the development of posibilities of image processing, stereoscopy, prices of web cameras and power of computers an opportunity to multiply an experience with working with 3D programs showed. From the picture from webcamera an estimation of a pose of user's head can be made. According to this pose a view on 3D scene can be changed. Then, when user moves his head, he will have a feeling as if monitor was a window through which one can see the scene behind. With the system which is the result of this project it will be possible to easily and cheaply add this kind of behaviour to any 3D application.

This thesis deals with the study of contactless and noninvasive methods for estimation of heart rate. Contactless measurement is based on capturing person faces by video camera and from sequences of pictures are estimated values of the heart rate. The theoretical part describes heart rate and methods that are being used to estimate heart rate from color changes in the face. It also contains testing of tracking algorithms. Practical part deals with user interface of program for contactless measurement of heart rate and its software solution. Thesis also contains statistical evaluation of program functionality.

One of the most important aspects for all life on this planet is theact to keep their cellular processes in a state where they do notreach equilibrium. One part in the upholding of this imbalanced stateis the barrier between the cells and their surroundings, created bythe cell membrane. In addition to experiments, the investigation ofprocesses occuring in the cell membrane can be performed by usingmolecular dynamics simulations. Through this method we can obtain anatomistic description of the dynamics associated with events that arenot accessible to experimental setups. Molecular dynamics relies onthe integration of Newton's equations of motion in order to sample therelevant parts of phase-space for the system, and therefore it isdependent on a correct description of the interactions between all thesimulated particles. In this thesis I first present an improved methodfor the calculation of long-range interactions in molecular dynamicssimulations, followed by a study of cholesterol's impact on thepermeation of small solutes across a lipid bilayer. The first paper presents a previously derived modification to theparticle-mesh Ewald method, which makes it possible to apply thisto long-range Lennard-Jones interactions. Old implementations of themethod have been haunted by an extreme performance degradation andhere I propose a solution to this problem by applying a modifiedinteraction potential. I further show that the historical treatmentof long-range interactions in simulations of lipid bilayers hasnon-negligible effects on their structural properties.In the second paper, this modification is improved such that the smallerrors introduced by the modified interaction potential becomenegligible. Furthermore, I demonstrate that I have also improved theimplementation of the method so that it now only incurs a performanceloss of roughly 15% compared to conventional simulations using theGromacs simulation package.The third paper presents a simulation study of cholesterol's effect onthe permeation of six different solutes across a variety of lipidbilayers. I analyze the effect of different head groups, tail lengths,and tail saturation by performing simulations of the solutes in fourdifferent bilayers, with cholesterol contents between 0% and50%. Analysis of the simulations shows that the impact of the surfacearea per lipid on the partitioning of the solute could be lower thanpreviously thought. Furthermore, a model with a laterallyinhomogeneous permeability in cholesterol-containing membranes isproposed, which could explain the large differences betweenpermeabilities from experiments and calculated partition coefficientsin simulations. / <p>QC 20140609</p>

In calculating the time evolution of an atomic system on diffusive timescales, off-lattice kinetic Monte Carlo (OLKMC) can sometimes be used to overcome the limitations of Molecular Dynamics. OLKMC relies on the harmonic approximation to Transition State Theory, in which the rate of rare transitions from one energy minimum to a neighboring minimum scales exponentially with an energy barrier on the potential energy surface. This requires locating the index-1 saddle point, commonly referred to as a transition state, that separates two neighboring energy minima. In modeling the evolution of an atomic system, it is desirable to find all the relevant transitions surrounding the current minimum. Due to the large number of minima on the potential energy surface, exhaustively searching the landscape for these saddle points is a challenging task. In examining the particular case of isolated Lennard-Jones clusters of around 50 particles, we observe very slow convergence of the total number of saddle points found as a function of successful searches. We seek to understand this behavior by modeling the distribution of successful searches and sampling this distribution to create a stochastic process that mimics this behavior. Finally, we will discuss an improvement to a rejection scheme for OLKMC where we terminate searches that appear to be failing early in the search process.

This study examines the prevalence of earnings management during the covid-19 pandemic in Swedish listed firms and aims to provide further evidence regarding earnings management practices during the pandemic. The study further investigates whether there are any differences between industries in terms of engaging in opportunistic accounting procedures. Previous research on earnings management during the covid-19 pandemic exclusively examines the year of 2020 as pandemic period. This study aims to contribute with further evidence by expanding the pandemic period and including the fiscal year of 2021, adding one additional fiscal year largely affected by the pandemic. Previous studies on the area finds contrasting results, where some studies find that firms tend to pursue income-increasing practices, while others deliberately decrease financial performance during a crisis. In order to perform this study, the modified Jones model is used to detect accrual-based earnings management through abnormal levels of discretionary accruals. The result of this study finds significantly increased levels of income-decreasing accrual-based earnings management, potentially suggesting that firms engaged in “big bath accounting” during the pandemic in order to boost financial performance in future periods. Additionally, the results show significantly increased levels of income-decreasing earnings management in 2/8 industries.

A novel method of calculating viscosity from molecular dynamics simulations is developed, benchmarked, and tested. The technique is a transient method which has the potential to reduce CPU requirements for many conditions. An initial sinusoidal velocity profile is overlaid upon the peculiar velocities of the individual molecules in an equilibrated simulation. The transient relaxation of this initial velocity profile is then compared to the corresponding analytical solution of the momentum equation by adjusting the viscosity-related parameters in the constitutive equation that relate the shear rate to the stress tensor. The newly developed Transient Molecular Dynamics (TMD) method was tested for a Lennard-Jones (LJ) fluid over a wide range of densities and temperatures. The simulated values were compared to an analytical solution of the boundary value problem for a Newtonian fluid. The resultant viscosities agreed well with those published for Equilibrium Molecular Dynamics (EMD) simulations up to a dimensionless density of 0.7. Application of a linear viscoelastic Maxwell constitutive equation was required to achieve good agreement at dimensionless densities greater than 0.7. When the Newtonian model is used for densities in the range of 0.1 to 0.3 and the Maxwell model is used for densities higher than 0.3, the TMD method was able to predict viscosities with an uncertainty of 10% or better. Application of the TMD method to multi-site molecules required the Jeffreys constitutive equation to adequately fit the simulation responses. TMD simulations were performed on model fluids representing n-butane, isobutane, n-hexane, water, methanol, and hexanol. Molecules with strong hydrogen bonding and Coulombic interactions agreed well with NEMD simulated values and experimental values. Simulated viscosities for nonpolar and larger molecules agreed with NEMD simulations at low to moderate densities, but deviated from these values at higher densities. These deviations are explainable in terms of potential model inaccuracies and the shear-rate dependence of both NEMD and TMD viscosity values. Results show that accurate viscosity predictions can be made for multi-site molecules as long as the shear-rate dependence of the viscosity is not too large or is adequately addressed.