In linguistics, words are usually considered to be composed of morphemes: units that carry meaning and are not further subdivisible. The task of this thesis is to create an automatic method for segmenting Czech words into morphemes, usable within the network of Czech derivational relations DeriNet. We created two different methods. The first one finds morpheme boundaries by differentiating words against their derivational parents, and transitively against their whole derivational family. It explicitly models morphophonological alternations and finds the best boundaries using maximum likelihood estimation. At worst, the results are slightly worse than the state of the art method Morfessor FlatCat, and they are significantly better in some settings. The second method is a neural network made to jointly predict segmentation and derivational parents, trained using the output of the first method and the derivational pairs from DeriNet. Our hypothesis that such joint training would increase the quality of the segmentation over training purely on the segmentation task seems to hold in some cases, but not in other. The neural model performs worse than the first one, possibly due to being trained on data which already contains some errors, multiplying them.

This thesis explores the experiences of managers who work flexibly. Flexible working policies are prevalent in all organizations in the UK because of the legislation giving specific groups of parents and carers the right to request flexible working. Many organizations extend the policies to all employees, yet the take-up is not as high as expected, particularly among staff at managerial levels. This thesis explores how managers construe and experience flexible working arrangements while successfully fulfilling their roles as managers of people. The exploratory study consisted of interviews with eight managers with unique flexible working patterns. Analysis of the interview transcripts identified concepts of consistency and adaptability. Consistency refers to meeting fixed needs from the work and non-work domains, and adaptability refers to the adjustment of schedules to meet the changing demands from those domains. The concepts of consistency and adaptability were further explored in the main study which is based on interviews with 24 women and 10 men who held managerial positions and had a flexible working arrangement which reduced their face time in the workplace. The research offers three main contributions to the literature. At a theoretical level, I propose a model which demonstrates how individuals use consistency and adaptability to meet the fixed and changing demands from the work and non-work domains. This model extends understanding of the complexity of the segmentation/integration continuum of boundary theory, explaining how and why managers use flexible working arrangements as a means of managing boundaries and achieving desired goals in both domains. Four distinct clusters emerged among the managerial participants in terms of the type and direction of adaptability, indicating the range of strategies used by managers to ensure the success of their flexible working arrangements. A detailed description of managers’ flexible working experiences is provided, adding to what is known about the role of manager through the exploration of the enactment of that role when working flexibly.

[pt] Este trabalho introduz o conceito de Dimensão Fractal de imagens, além de fazer uma aplicação de alguns métodos usuais de segmentação, visando o controle automático de falhas em tecidos têxteis. Foi desenvolvido um sistema dedicado à indústria têxtil objetivando a detecção de possíveis falhas. Uma indústria têxtil se particulariza por ter produção contínua. A característica planar dos produtos finais deste tipo de produção torna, de uma forma geral, inviável a utilização das técnicas de extração de características morfológicas ou geométricas, usualmente empregadas em sistemas de controle de qualidade baseados na visão. Basicamente o sistema implementado compara informações obtidas de imagens digitalizadas. Estas informações variam de acordo com o método escolhido que pode ser um dos dois grupos: métodos de segmentação e dimensão fractal. Para implementação no sistema, métodos de segmentação conhecidos foram adaptados e aperfeiçoados visando a determinação de variações em uma imagem do produto (o que caracteriza a existência de uma falha). Na utilização da Dimensão Fractal como uma ferramenta para análise de imagens e controle de qualidade utiliza-se um algoritmo eficiente. Este algoritmo calcula os valores de dimensão fractal de imagens em toda a região teoricamente admissível (2 menor ou igual DF menor ou igual 3). O sistema implementado foi experimentado em tecidos planos, onde os vários métodos foram comparados quanto a sua eficiência, precisão e aplicabilidade. / [en] This work presents na application of Fractal Dimension and Segmentation techniques for Automated Visual Inspection. A System was developed for textile industries that aims automatic failure detection. Such as birch wood board and steel slabs industries, this environment has particular characteristics in which morphological feature extraction can not be used for visual quality control. The System compare data of the digital images. These images data depend on the selected method. Two methods can be used: segmentation or fractal dimension. Segmentation techniques were adapted and improved in order to detect variations of the product s images (signal of defect). The image analysis by the use of fractal dimension is based on a efficient algorithm. This algorithm calculate the fractal dimension images values throughout the range between two and three (2 less than or equal to DF less than or equal to 3). The system have been tested using textile images and the techniques were compared by its efficiency, accuracy and application.

Optimization methods have been widely utilized in the field of imaging science, such as image denoising, image segmentation, image contrast adjustment, high dynamic rang imaging, etc. In recent decades, it is becoming more and more popular to re- formulate an image processing problem into an energy minimization problem, then solve for the minimizer by some optimization based methods. In this thesis, we concern solving three popular issues in image processing and computational photography by optimization based methods, which are image segmentation, bit-depth expansion, and high dynamic range image tone mapping problems. The contribution of this thesis can be illustrated in three parts separately according to different topics. For the image segmentation problem, we present a multi-phase image segmentation model based on the histogram of the Gabor feature space, which consists of responses from a set of Gabor filters with various orientations, scales and frequencies. Our model replaces the error function term in the original fuzzy region competition model with squared 2-Wasserstein distance function, which is a metric to measure the distance of two histogram. The energy functional is minimized by alternating direction method of multiplier, and the existence of the closed-form solutions is guaranteed when the exponent of the fuzzy membership term being 1 or 2. The experimental results show the advantage of our proposed method compared to other recent methods. As for the bit-depth expansion problem, we develop a variational approach containing an energy functional to determine a local mapping function for bit-depth expansion via a smoothing technique, such that each pixel can be adjusted locally to a high bit-depth value. In order to enhance the contrast of the low bit-depth images, we make use of the histogram equalization technique for such local mapping function. Both bit-depth expansion and histogram equalization terms can be combined together into the resulting objective function. In order to minimize the differences among the local mapping functions at the nearby pixel locations, the spatial regularization of the mapping is incorporated in the objective function. Regarding the tone mapping problem for high dynamic range images, we pro- pose a computational tone mapping operator which makes use of a localized gamma correction. Our tone mapping operator combines the two subproblems in the tone mapping problem, i.e. luminance compression and color rendering, into one general framework. The bright regions and dark regions can be distinguished and treated differently. In our method, we propose two adjustment rules according to the perceptual preference of human visual system towards contrast and colors respectively. The resulting tone mapped images have a natural looking and the highest score in our observer subjective test. Based on the motivation of our computational tone mapping operator, we propose a variational method for image tone mapping problem. The core idea is to minimize the difference of the local contrast between the tone mapped image and the high dynamic range image under some constraints. The energy functional contains a local contrast fidelity term and a L-2 total variation regularization term. Local gamma correction is also applied as our previous computational model and the unknown variables are the non-uniform gamma values. The non-uniform gamma values for each pixel can be obtained by minimizing the fidelity term, while the smoothing term ensures the gamma values for nearby pixels not varying too much from each other. The results by both our computational and variational tone mapping operators show advantage in preserving the detailed image contents in the bright and dark regions. Keywords: optimization, alternating direction method of multipliers, variational model, image segmentation, Mumford-Shah model, Gabor filter, contrast adjustment, histogram equalization, bit-depth expansion, dynamic range, HDR imaging, tone mapping operators, gamma correction, color rendering.

Esta pesquisa é resultado do reconhecimento, no mercado hoteleiro, de algumas vantagens em se conhecer o conjunto de necessidades específicas de um determinado tipo de hóspede. O turismo de negócios é visto como um importante segmento do mercado turístico e como forte gerador da demanda hoteleira em algumas regiões. A proposta é, a partir da apresentação de características do mercado turístico e do setor hoteleiro, discutir a percepção do turista de negócios quanto aos atributos oferecidos pelos hotéis que justificam a preferência por dado empreendimento para sua estadia. A pesquisa bibliográfica, aliada a um estudo exploratório na cidade de São Paulo, permite algumas conjecturas a respeito do comportamento do turista de negócios no que diz respeito à hospedagem. Em virtude do dinamismo desta demanda, espera-se incentivar com esta pesquisa a continuidade, por profissionais e acadêmicos, dos estudos a respeito do mercado hoteleiro para o turismo de negócios. / business tourist behavior concerning hosting. Due to the dynamism of hotel demand, is expected that professionals and academics be stimulated to prospect and study the hotel market for business tourism. This research is the outcome of the recognition, in the hotel market, of the advantages of knowing the needs of a certain guest profile. Business tourism is an important segment of the tourism market and generates a great demand in some specific regions. The purpose of the research is, from the overview of the characteristics of tourism market and hotels segment, discuss the business tourist perspective of the attributes offered by hotels that justify their preference for a specific project. Research bibliography combined with an exploration in the city of São Paulo, allowed some hypothesis regarding

Fast, automated segmentation of the thalamic nuclei in the brain has long been desired as it provides for direct visualization of the target for certain procedures like Deep Brain Stimulation (DBS) that target a specific nucleus. It is also beneficial in the study of other pathologies that pertain to different nuclei. In this thesis, a novel approach to fast automated segmentation of thalamic nuclei called Shortened Template and THalamus for Optimal Multi Atlas Segmentation (ST THOMAS) was developed using the multi-atlas segmentation approach. It was designed with a focus on robustness and speed by making use of an averaged template for registration and cropping the inputs and the template. The performance of ST THOMAS was first evaluated on 7T MRI data by comparing with manual delineation (ground truth) by an expert neuroradiologist. Dice coefficients and Volumetric Similarity Indices were used as metrics. To extend the applicability of this method, 3T MRI data were also evaluated. Finally, applications to real time ventralintermideiate (VIM) nucleus targeting for DBS and study of the effects of alcoholism are demonstrated.

First generation image coding standards are now well-established and coders based on these standards are commercially available. However, for emerging applications, good quality at even lower bitrates is required. Ways of exploiting higher level visual information are currently being explored by the research community in order to achieve high compression. Unfortunately very high level approaches are bound to be restrictive as they are highly dependent on the accuracy of lower-level vision operations. Region-based coding only relies on mid-level image processing and thus is viewed as a promising strategy. In this work, substantial advances to the field of region-based video compression are made by considering the complete scheme. Thus, improvements to the failure regions coding and the motion compensation components have been devised. The failure region coding component was improved by predicting the texture inside the failure region from the neighbourhood of the region. A significant gain over widely used techniques such as the SA-DCT was obtained. The accuracy of the motion compensation component was increased by keeping an accurate internal representation for each region both at the encoder and the decoder side. The proposed region-based coding system is also evaluated against other systems, including the MPEG4 codec which has been recently approved by the MPEG community.

The large amount of data accumulated by applications is stored in a database. Because of the large amount, name conflicts or missing values sometimes occur. This prevents certain types of analysis. In this work, we solve the name conflict problem by comparing the similarity of the data, and changing the test data into the form of a given template dataset. Studies on data use many methods to discover knowledge from a given dataset. One popular method is association rules mining, which can find associations between items. This study unifies the incomplete data based on association rules. However, most rules based on traditional association rules mining are item-to-item rules, which is a less than perfect solution to the problem. The data recovery system is based on complex association rules able to find two more types of association rules, prefix pattern-to-item, and suffix pattern-to-item rules. Using complex association rules, several missing values are filled in. In order to find the frequent prefixes and frequent suffixes, this system used FP-tree to reduce the time, cost and redundancy. The segment phrases method can also be used for this system, which is a method based on the viscosity of two words to split a sentence into several phrases. Additionally, methods like data compression and hash map were used to speed up the search.

À l’ère du numérique, valoriser les données en leur donnant un sens est un enjeu capital pour supporter la prise de décision stratégique et cela dans divers domaines, notamment dans le domaine du marketing numérique ou de la santé, ou encore, dans notre contexte, pour une meilleure compréhension de la biologie des structures des acides nucléiques. L’un des défis majeurs de la biologie structurale concerne l’étude des structures des acides ribonucléiques (ARN), les effets de ces structures et de leurs altérations sur leurs fonctions. Contribuer à cet enjeu important est l’objectif de cette thèse. Celle-ci s’inscrit principalement dans le développement de méthodes et d’outils pour l’exploration efficace des structures secondaires d’ARN. En effet, explorer les structures secondaires d’ARN contribue à lever le voile sur leur fonction et permet de mieux cerner leur implication spécifique au sein des processus cellulaires. Dans ce contexte nous avons développé le modèle des super-n-motifs qui contribue à une meilleure représentation de la complexité structurale des ARN et offre un moyen efficace d’évaluer la similarité des structures d’ARN en tenant compte de cette complexité. Le modèle des super-n-motifs facilite l’étude des ARN dont le rôle est inconnu. Il permet de poser des hypothèses sur la ou les fonctions des ARN lorsque ceux-ci partagent une similarité structurale sans équivoque. Nous avons aussi développé la plateforme structurexplor pour faciliter l’exploration des structures secondaires, c’est-à-dire de permettre, en quelques clics, de caractériser les populations de structures d’ARN en, par exemple, faisant ressortir les groupes d’ARN partageant des structures similaires. La mise en œuvre du modèle des super-n-motifs et de la plateforme structurexplor a contribué à une meilleure compréhension de la phylogénie structurale des viroïdes qui sont des agents pathogènes à ARN attaquant les plantes, phylogénie jusqu’alors basée que sur leurs séquences.

Medical image analysis is a ubiquitous and essential part of modem health care. A crucial first step to this is segmentation, which is often complicated by many factors including subject diversity, pathology, noise corruption, and poor image resolution. Traditionally, manual tracing by experts was done. While considered accurate, this process is time consuming and tedious, especially when performed slice-by-slice on three-dimensional (3D) images over large datasets or on two-dimensional (2D) but topologically complicated images such as a retinography. On the other hand, fully-automated methods are typically faster, but work best with data-dependent, carefully tuned parameters and still require user validation and refinement. This thesis contributes to the field of medical image segmentation by proposing a highly-automated, interactive approach that effectively merges user knowledge and efficient computing. To this end, our work focuses on graph-based methods and offer globally optimal solutions. First, we present a novel method for 3D segmentation based on a 3D Livewire approach. This approach is an extension of the 2D Livewire framework, and this method is capable of handling objects with large protrusions, concavities, branching, and complex arbitrary topologies. Second, we propose a method for efficiently segmenting 2D vascular networks, called ‘Live-Vessel’. Live-Vessel simultaneously extracts vessel centrelines and boundary points, and globally optimizes over both spatial variables and vessel radius. Both of our proposed methods are validated on synthetic data, real medical data, and are shown to be highly reproducible, accurate, and efficient. Also, they were shown to be resilient to high amounts of noise and insensitive to internal parameterization. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate