During the course of this thesis, two scenarios are considered. In the first one, we contribute to feature extraction algorithms. In the second one, we use features to improve object detection solutions and localization. The two scenarios give rise to into four thesis sub-goals. First, we present a new shape skeleton pruning algorithm based on contour approximation and the integer medial axis. The algorithm effectively removes unwanted branches, conserves the connectivity of the skeleton and respects the topological properties of the shape. The algorithm is robust to significant boundary noise and to rigid shape transformations. It is fast and easy to implement. While shape-based solutions via boundary and skeleton analysis are viable solutions to object detection, keypoint features are important for textured object detection. Therefore, we present a keypoint featurebased planar object detection framework for vision-based localization. We demonstrate that our framework is robust against illumination changes, perspective distortion, motion blur, and occlusions. We increase robustness of the localization scheme in cluttered environments and decrease false detection of targets. We present an off-line target evaluation strategy and a scheme to improve pose. Third, we extend planar object detection to a real-time approach for 3D object detection using a mobile and uncalibrated camera. We develop our algorithm based on two novel naive Bayes classifiers for viewpoint and feature matching that improve performance and decrease memory usage. Our algorithm exploits the specific structure of various binary descriptors in order to boost feature matching by conserving descriptor properties. Our novel naive classifiers require a database with a small memory footprint because we only store efficiently encoded features. We improve the feature-indexing scheme to speed up the matching process creating a highly efficient database for objects. Finally, we present a model-free long-term tracking algorithm based on the Kernelized Correlation Filter. The proposed solution improves the correlation tracker based on precision, success, accuracy and robustness while increasing frame rates. We integrate adjustable Gaussian window and sparse features for robust scale estimation creating a better separation of the target and the background. Furthermore, we include fast descriptors and Fourier spectrum packed format to boost performance while decreasing the memory footprint. We compare our algorithm with state-of-the-art techniques to validate the results.

The subject of this diploma thesis is a dynamic analysis of phenomena occurring in the thin cylindrical region, ie in a gab, which flows liquid by the given parameters. Through the simplifications introduced, it comes to the problem of structural and forced vibration for both the horizontal pulsation in a tube and the area of a compensating piston. The described method uses the separation of variables for pressure function. The speed function is derived from the already known pressure field distribution using the Navier-Stokes equation.

This diploma thesis deals with the issue of operational modal analysis, which is one of the newer areas of structural dynamics that makes it possible to estimate modal properties of structures directly during their operation. The aim of this work is to create theoretical basis of operational modal analysis, more specifically of Frequeny Domain Decomposition method. Furthermore, apply this method to a selected structure and verify the achieved results using computational modelling and experimental modal analysis. The technique of dividing the experiment to separate measurements using reference sensor was used. Evaluation of the operational modal analysis was performed using self-created function for the MATLAB software and using the commercial PULSE Operational Modal Analysis software. Obtained results were verified by computational modelling and experimental modal analysis.

The presented diploma thesis deals with the modeling of the rotor packet in dynamic calculations and analysis of the influence of the packing pressing force on the critical speed of the electric motor. The thesis can be divided into three main parts. The first part discusses the current level of rotor packet modeling in dynamic calculations and also deals with analytical theories for subsequent experimental tests. The second part describes the implementation of two experiments to determine the equivalent modulus of elasticity. The first experiment is a four-point bending test and the second is an experimental modal analysis. The third part is used for comparing the results obtained from experimental tests and describes the calculation of the critical speed of the rotor package pressed on the shaft by the finite element method.

In this thesis is wrote up a complete procedure of stator core stiffness determination using finite element method (FEM), subsequently, the stator core stiffness database was created according to this procedure. In the next chapter is carried out a sensitivity analysis of one chosen stator core, which led to identifying parameters that affected its stiffness the most. Finally, a modal analysis of one synchronous machine was performed using FEM, in order to determine the effect of stator core stiffness on modal characteristics of the machine. Several modal shapes and corresponding natural frequencies were computed in this analysis, then only a few shapes with the highest core effect were chosen and compared.

This thesis focuses on creation of two models of high-rise construction - beam and shell. A dynamic analysis is performed on these models. The selected program is RFEM software and the created high-rise construction is Hearst Tower skyscraper, which is interesting for its diagrid system around its perimeter. There is described process of creating both models and associated problems. Dynamic analysis is performed using standard EN 1998-1. There belongs calculations of sufficient number of modal shapes, spectral analysis and time analysis. Then the results are compared on selected point on the both models.

Les déchets plastiques s'accumulent depuis plusieurs décennies dans les océans où ils se fragmentent en particules appelés microplastiques lorsque leur taille est inférieure à 5 mm. Ces microplastiques sont retrouvés dans toutes les eaux du globe, dans les sédiments ainsi que dans de nombreux organismes marins. Le devenir physicochimique à long terme de ces particules et leur possible fragmentation en nanoplastiques sont complexes, encore peu documentés et nécessitent des études en laboratoire.Afin de comprendre les processus liés à la photodégradation et à la fragmentation des polymères dans l’environnement, mais également dans le but d’'appréhender l’évolution des fragments générés au cours de l’irradiation, un protocole de vieillissement accéléré en milieu abiotique a été mis en place sur des polymères modèles. Le suivi de l’oxydation et de la fragmentation des deux polymères étudiés,polyéthylène basse densité et polypropylène, a été mené à l’aide de techniques spectroscopiques (infrarouge, Raman), DSC, angles de contact, et microscopiques (lumière polarisée, MEB, AFM…).Ce travail a permis de mettre en évidence l’influence significative de l’environnement et de la morphologie initiale des polymères sur leurs cinétiques de vieillissement et leurs mécanismes de fissuration. Ainsi des distributions en nombres, tailles et formes de fragments très différentes ont été obtenues pour les deux polymères selon la présence d’eau. Enfin, après un long temps d’irradiation, des produits de dégradation ont pu être détectés mais la production significative de nanoplastiques n’a pas été démontrée. L'hypothèse d'une taille limite de fragmentation devrait être envisagée. / Plastic wastes have been accumulating for several decades in the oceans where they break up into particles called microplastics when their size is less than 5 mm. These microplastics are found in all earth’s waters, in sediments and in many marine organisms. Their long-term physico-chemical fate and their possible fragmentation into nanoplastics are complex, still poorly documented and require laboratory studies.In order to understand the processes related to photodegradation and fragmentation of polymers, but also in order to understand the evolution of these fragments during irradiation, an accelerated aging protocol in abiotic conditions has been set up. The oxidation and fragmentation of two model polymers, low density polyethylene and polypropylene, were monitored using spectroscopic techniques (InfraRed, Raman), DSC, contact angles and microscopic technics (light microscopy, polarized light, SEM, AFM ...).This work has demonstrated a significant influence of the environment and the initial morphology of the polymers on their kinetics of aging and cracking mechanisms. This lead to significantly different distributions in numbers, sizes and shapes of the generated fragments. Moreover, after a long time of irradiaiton, other degradation products could be detected but the significant production of nanoplastics has not been demonstrated. The possibility of a size limit below which the fragmentation rate of plastics would strongly decrease should be considered.

In this paper a new equivalent circuit is presented which describes the dynamics of a prototype micro-gyro sensor. The concept takes advantage of the principles employed in vibratory gyro sensors and the ductile attributes of GalFeNOL to target high sensitivity and shock tolerance. The sensor is designed as a tuning fork structure. A GalFeNOL patch attached to the y-z surface of the drive prong causes both prongs to bending the x-z plane (about the y axis) and a patch attached to the x-z surface of the sensing prong detects Coriolis-force induced bending in the y-z plane (about the x axis). A permanent magnet is bonded on top of each prong to give bias magnetic fields. A solenoid coil surrounding the drive prong is used to produce bending in the x-z plane of both prongs. The sensing prong is surrounded by a solenoid coil with N turns in which a voltage proportional to the time rate of change of magnetic flux is induced. The equivalent circuit enables the efficient modeling of a gyro sensor and an electromechanical behavioral simulation using the circuit simulator SPICE. The prongs are modeled as wave guiding bending beams which are coupled to the electromagnetic solenoid coil transducer. In contrast to known network approaches, the proposed equivalent circuit is the first tuning fork model, which takes full account of the fictitious force in a constant rotating frame of reference. The Coriolis force as well as the centrifugal force on a concentrated mass are considered.

In pieces I wanted to create a design solution inspired by traditional wooden furnitures. Furnitures, like the kitchen sofa with built in storage and a place to sleep or the lockable free-standing linen cabinet that held hand crafted treasures safe and clean inside. Furnitures that are large in size and can be hard to fit into smaller homes today but that holds a huge sentimental value to the owner and are often handed down through generations as heirlooms.   So what if I could weave in a piece of my own family history into a design solution that can fit in smaller living spaces while still be an attractive choice for larger ones?  Pieces resulted in two wooden cabinets that tell part of my family history through the repetitative, organic shapes that I have implemented into them. They are a tribute to my grandmother Aino and her three year younger sister Hanna and my way of symbolically reclaiming their right to the land, forests and lakes around their childhood home that the two sisters didn´t get to share with their siblings in the inheritance after my great grandparents.  The methods I have used in my research are surveys, workshops, in-depth interviews and my own observations. Major design decisions have been based on the input from my research process with the goal to display two full scale cabinets at Konstfack's spring exhibition in May 2023.

further development of the simple and promising analytical probabilistic approach / Urban stormwater management aims at mitigating the adverse impacts of urbanization. Hydrological models are used in support of stormwater management planning and design. There are three main approaches that can be applied for this modeling purpose: (1) continuous simulation approach which is accurate but time-consuming; (2) design storm approach, which is widely used and its accuracy highly depends on the selected antecedent moisture conditions and temporal distribution of design storms; and (3) the analytical probabilistic approach which is recently developed and still not used in practice. Although it is time-effective and it can produce results as accurate as the other two approaches; the analytical probabilistic approach requires further developments in order to make it more reliable and accurate. For this purpose, three subtopics are investigated in this thesis. (1) Rainfall data analysis as required by the analytical probabilistic approach with emphasis on testing the exponentiality of rainfall event duration, volume and interevent time (i.e., time separating it from its preceding rainfall event). A goodness-of-fit testing procedure that is suitable for this kind of data analysis was proposed. (2) Derivation of new analytical probabilistic models for peak discharge rate incorporating trapezoidal and triangular hydrograph shapes in order to include all possible catchment’s responses. And (3) the infiltration process is assumed to continue until the end of the rainfall event; however, the soil may get saturated earlier and the excess amount would contribute to the runoff volume which may have adverse impact if not taken into consideration. Thus, in addition to the infiltration process, the saturation excess runoff is also included and new models for flood frequencies are developed. All the models developed in this thesis are tested and compared to methods used in practice, reasonable results were obtained. / Thesis / Doctor of Philosophy (PhD) / Urban stormwater management aims at mitigating the adverse impacts of urbanization. Hydrological models are used in support of stormwater management planning and design. The analytical probabilistic stormwater management model (APSWM) is a promising tool for planning and design analysis. The purpose of this thesis is to further develop APSWM in order to make it more reliable and accurate. First, a clear procedure for rainfall data analysis as required by APSWM is provided. Second, a new APSWM is derived incorporating other runoff temporal-distribution patterns. Finally, the possibility of soil layer saturation while it is still raining is added to the model. All the models developed in this thesis are tested and compared to methods used in engineering practice, reasonable results were obtained.