Дисертації з теми "Segmentation models"
Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Segmentation models.
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Segmentation models".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
Camilleri, Liberato. "Statistical models for market segmentation." Thesis, Lancaster University, 2005. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.441119.
Повний текст джерела
2
Heiler, Matthias. "Image models for segmentation and recognition." [S.l. : s.n.], 2006. http://madoc.bib.uni-mannheim.de/madoc/volltexte/2006/1306.
Повний текст джерела
3
Eslami, Seyed Mohammadali. "Generative probabilistic models for object segmentation." Thesis, University of Edinburgh, 2014. http://hdl.handle.net/1842/8898.
Повний текст джерелаАнотація:
One of the long-standing open problems in machine vision has been the task of ‘object segmentation’, in which an image is partitioned into two sets of pixels: those that belong to the object of interest, and those that do not. A closely related task is that of ‘parts-based object segmentation’, where additionally each of the object’s pixels are labelled as belonging to one of several predetermined parts. There is broad agreement that segmentation is coupled to the task of object recognition. Knowledge of the object’s class can lead to more accurate segmentations, and in turn accurate segmentations can be used to obtain higher recognition rates. In this thesis we focus on one side of this relationship: given the object’s class and its bounding box, how accurately can we segment it? Segmentation is challenging primarily due to the huge amount of variability one sees in images of natural scenes. A large number of factors combine in complex ways to generate the pixel intensities that make up any given image. In this work we approach the problem by developing generative probabilistic models of the objects in question. Not only does this allow us to express notions of variability and uncertainty in a principled way, but also to separate the problems of model design and inference. The thesis makes the following contributions: First, we demonstrate an explicit probabilistic model of images of objects based on a latent Gaussian model of shape. This can be learned from images in an unsupervised fashion. Through experiments on a variety of datasets we demonstrate the advantages of explicitly modelling shape variability. We then focus on the task of constructing more accurate models of shape. We present a type of layered probabilistic model that we call a Shape Boltzmann Machine (SBM) for the task of modelling foreground/background (binary) and parts-based (categorical) shapes. We demonstrate that it constitutes the state-of-the-art and characterises a ‘strong’ model of shape, in that samples from the model look realistic and that it generalises to generate samples that differ from training examples. Finally, we demonstrate how the SBM can be used in conjunction with an appearance model to form a fully generative model of images of objects. We show how parts-based object segmentations can be obtained simply by performing probabilistic inference in this joint model. We apply the model to several challenging datasets and find that its performance is comparable to the state-of-the-art.
4
Li, Zhi. "Variational image segmentation, inpainting and denoising." HKBU Institutional Repository, 2016. https://repository.hkbu.edu.hk/etd_oa/292.
Повний текст джерелаАнотація:
Variational methods have attracted much attention in the past decade. With rigorous mathematical analysis and computational methods, variational minimization models can handle many practical problems arising in image processing, such as image segmentation and image restoration. We propose a two-stage image segmentation approach for color images, in the first stage, the primal-dual algorithm is applied to efficiently solve the proposed minimization problem for a smoothed image solution without irrelevant and trivial information, then in the second stage, we adopt the hillclimbing procedure to segment the smoothed image. For multiplicative noise removal, we employ a difference of convex algorithm to solve the non-convex AA model. And we also improve the non-local total variation model. More precisely, we add an extra term to impose regularity to the graph formed by the weights between pixels. Thin structures can benefit from this regularization term, because it allows to adapt the weights value from the global point of view, thus thin features will not be overlooked like in the conventional non-local models. Since now the non-local total variation term has two variables, the image u and weights v, and it is concave with respect to v, the proximal alternating linearized minimization algorithm is naturally applied with variable metrics to solve the non-convex model efficiently. In the meantime, the efficiency of the proposed approaches is demonstrated on problems including image segmentation, image inpainting and image denoising.
5
Yeo, Si Yong. "Implicit deformable models for biomedical image segmentation." Thesis, Swansea University, 2011. https://cronfa.swan.ac.uk/Record/cronfa42416.
Повний текст джерелаАнотація:
In this thesis, new methods for the efficient segmentation of images are presented. The proposed methods are based on the deformable model approach, and can be used efficiently in the segmentation of complex geometries from various imaging modalities. A novel deformable model that is based on a geometrically induced external force field which can be conveniently generalized to arbitrary dimensions is presented. This external force field is based on hypothesized interactions between the relative geometries of the deformable model and the object boundary characterized by image gradient. The evolution of the deformable model is solved using the level set method so that topological changes are handled automatically. The relative geometrical configurations between the deformable model and the object boundaries contributes to a dynamic vector force field that changes accordingly as the deformable model evolves. The geometrically induced dynamic interaction force has been shown to greatly improve the deformable model performance in acquiring complex geometries and highly concave boundaries, and give the deformable model a high invariance in initialization configurations. The voxel interactions across the whole image domain provides a global view of the object boundary representation, giving the external force a long attraction range. The bidirectionality of the external force held allows the new deformable model to deal with arbitrary cross-boundary initializations, and facilitates the handling of weak edges and broken boundaries. In addition, it is shown that by enhancing the geometrical interaction field with a nonlocal edge-preserving algorithm, the new deformable model can effectively overcome image noise. A comparative study on the segmentation of various geometries with different topologies from both synthetic and real images is provided, and the proposed method is shown to achieve significant improvements against several existing techniques. A robust framework for the segmentation of vascular geometries is described. In particular, the framework consists of image denoising, optimal object edge representation, and segmentation using implicit deformable model. The image denoising is based on vessel enhancing diffusion which can be used to smooth out image noise and enhance the vessel structures. The image object boundaries are derived using an edge detection technique which can produce object edges of single pixel width. The image edge information is then used to derive the geometric interaction field for optimal object edge representation. The vascular geometries are segmented using an implict deformable model. A region constraint is added to the deformable model which allows it to easily get around calcified regions and propagate across the vessels to segment the structures efficiently. The presented framework is ai)plied in the accurate segmentation of carotid geometries from medical images. A new segmentation model with statistical shape prior using a variational approach is also presented in this thesis. The proposed model consists of an image attraction force that propagates contours towards image object boundaries, and a global shape force that attracts the model towards similar shapes in the statistical shape distribution. The image attraction force is derived from gradient vector interactions across the whole image domain, which makes the model more robust to image noise, weak edges and initializations. The statistical shape information is incorporated using kernel density estimation, which allows the shape prior model to handle arbitrary shape variations. It is shown that the proposed model with shape prior can be used to segment object shapes from images efficiently.
6
Barker, Simon A. "Image segmentation using Markov random field models." Thesis, University of Cambridge, 1998. https://www.repository.cam.ac.uk/handle/1810/272037.
Повний текст джерела
7
Barker, S. A. "Unsupervised image segmentation using Markov Random Field models." Thesis, University of Cambridge, 1999. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.596368.
Повний текст джерелаАнотація:
The development of a fully unsupervised algorithm to achieve image segmentation is the central theme of this dissertation. Existing literature falls short of such a goal providing many algorithms capable of solving a subset of this highly challenging problem. Unsupervised segmentation is the process of identifying and locating the constituent regions of an observed image, while having no prior knowledge of the number of regions. The problem can be formulated in a Bayesian framework and through the use of an assumed model unsupervised segmentation can be posed as a problem of optimisation. This is the approach pursued throughout this dissertation. Throughout the literature, the commonly adopted model is an hierarchical image model whose underlying components are various forms of Markov Random Fields. Gaussian Markov Random Field models are used to model the textural content of the observed image's regions, while a Potts model provides a regularisation function for the segmentation. The optimisation of such highly complicated models is a topic that has challenged researchers for several decades. The contribution of this thesis is the introduction of new techniques allowing unsupervised segmentation to be carried using a single optimisation process. It is hoped that these algorithms will facilitate the future study of hierarchical image models and in particular the discovery of further models capable of more closely fitting real world data. The extensive literature surrounding Markov Random Field models and their optimisation is reviewed early in this dissertation, as is the literature concerning the selection of features to identify the textural content of an observed image. In the light of these reviews new algorithms are proposed that achieve a fusion between concepts originating in both these areas. Algorithms previously applied in statistical mechanics form an important part of this work. The use of various Markov Chain Monte Carlo algorithms is prevalent and in particular, the reversible jump sampling algorithm is of great significance. It is the combination of several of these algorithms to form a single optimisation framework that lies at the heart of the most successful algorithms presented here.
8
Shepherd, T. "Dynamical models and machine learning for supervised segmentation." Thesis, University College London (University of London), 2009. http://discovery.ucl.ac.uk/18729/.
Повний текст джерелаАнотація:
This thesis is concerned with the problem of how to outline regions of interest in medical images, when the boundaries are weak or ambiguous and the region shapes are irregular. The focus on machine learning and interactivity leads to a common theme of the need to balance conflicting requirements. First, any machine learning method must strike a balance between how much it can learn and how well it generalises. Second, interactive methods must balance minimal user demand with maximal user control. To address the problem of weak boundaries,methods of supervised texture classification are investigated that do not use explicit texture features. These methods enable prior knowledge about the image to benefit any segmentation framework. A chosen dynamic contour model, based on probabilistic boundary tracking, combines these image priors with efficient modes of interaction. We show the benefits of the texture classifiers over intensity and gradient-based image models, in both classification and boundary extraction. To address the problem of irregular region shape, we devise a new type of statistical shape model (SSM) that does not use explicit boundary features or assume high-level similarity between region shapes. First, the models are used for shape discrimination, to constrain any segmentation framework by way of regularisation. Second, the SSMs are used for shape generation, allowing probabilistic segmentation frameworks to draw shapes from a prior distribution. The generative models also include novel methods to constrain shape generation according to information from both the image and user interactions. The shape models are first evaluated in terms of discrimination capability, and shown to out-perform other shape descriptors. Experiments also show that the shape models can benefit a standard type of segmentation algorithm by providing shape regularisers. We finally show how to exploit the shape models in supervised segmentation frameworks, and evaluate their benefits in user trials.
9
Ljolje, A. "Intonation and phonetic segmentation using hidden Markov models." Thesis, University of Cambridge, 1986. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.377219.
Повний текст джерела
10
Chalana, Vikram. "Deformable models for segmentation of medical ultrasound images /." Thesis, Connect to this title online; UW restricted, 1996. http://hdl.handle.net/1773/8025.
Повний текст джерела
11
Chaieb, Ines. "Essays on international asset pricing under segmentation and PPP deviations." Thesis, McGill University, 2006. http://digitool.Library.McGill.CA:80/R/?func=dbin-jump-full&object_id=102485.
Повний текст джерелаАнотація:
This dissertation comprises two essays. The first essay develops and tests a theoretical model that provides new insights when markets are partially segmented and the purchasing power parity (PPP) is violated which seems to be the case for the majority of national markets. The theoretical part derives closed form solutions for asset prices and portfolio holdings. Particularly, we show that deviations from PPP in mildly segmented markets induce a new form of systematic risk, termed segflation risk, and in equilibrium investors require compensation for this risk. A strong feature of the model is that it provides a theoretical framework for testing important issues; such as, pricing of foreign exchange risk and world market structure. The model also nests several existing international asset pricing models and thus provides a framework to distinguish empirically between competing models. The empirical part of the essay provides an empirical validation of the model for eight major emerging markets. The results give support to the model and point to the importance of the segflation risk which is statistically and economically significant.The second essay uses our theoretical model to address the question of whether the IFC investable indices are priced globally or locally. Indeed S&P/IFC provides two emerging market indices: the IFC global index (IFCG) and its subset the IFC investable index (IFCI). Since the IFCI is fully investable, both the academic and practitioners implicitly assume that this subset of emerging markets is priced in the global context. This is a critical assumption for corporate finance decisions and portfolio management. Hence, this essay investigates the pricing behavior of the IFCI index returns using a conditional version of our model that allows for segmentation and PPP deviations. The results suggest that local factors are important in explaining returns of the IFC investable indices and that the return behavior of IFCI indices is similar to that of the IFCG.
12
Chen, Liyuan. "Variational approaches in image recovery and segmentation." HKBU Institutional Repository, 2015. https://repository.hkbu.edu.hk/etd_oa/227.
Повний текст джерелаАнотація:
Image recovery and segmentation are always the fundamental tasks in image processing field, because of their so many contributions in practical applications. As in the past ten years, variational methods have achieved a great success on these two issues, in this thesis, we continue to work on proposing several new variational approaches for restoring and segmenting an image. This thesis contains two parts. The first part addresses recovering an image and the second part emphasizes on segmenting. Along with the wide utilization of magnetic resonance imaging (MRI) technique, we particularly deal with blurry images corrupted by Rician noise. In chapter 1, two new convex variational models for recovering an image corrupted by Rician noise with blur are presented. These two models are motivated by the non-convex maximum-a-posteriori (MAP) model proposed in the prior papers. In the first method, we use an approximation item to the zero order of the modified Bessel function in the MAP model and add an entropy-like item to obtain a convex model. Through studying on the statistical properties of Rician noise, we bring up a strictly convex model by adding an additional data-fidelity term in the MAP model in the second method. Primal-dual methods are applied to solve the models. The simulation outcomes show that our models outperform some existed effective models in both recovery image quality and computational time. Cone beam CT (CBCT) is routinely applied in image guided radiation therapy (IGRT) to help patient setup. Its imaging dose, however, is still a concern, limiting its wide applications. It has been an active research topic to develop novel technologies for radiation dose reduction. In chapter 2, we propose an improvement of practical CBCT dose control scheme - temporal non-local means (TNLM) scheme for IGRT. We denoise the scanned image with low dose by using the previous images as prior knowledge. We combine deformation image registration and TNLM. Different from the TNLM, in the new method, for each pixel, the search range is not fixed, but based on the motion vector between the prior image and the obtained image. By doing this, it is easy to find the similar pixels in the previous images, but also can reduce the computational time since it does not need large search windows. The phantom and patient studies illuminate that the new method outperforms the original one in both image quality and computational time. In the second part, we present a two-stage method for segmenting an image corrupted by blur and Rician noise. The method is motivated by the two-stage segmentation method developed by the authors in 2013 and restoration method for images with Rician noise. First, based on the statistical properties of Rician noise, we present a new convex variant of the modified Mumford-Shah model to get the smooth cartoon part {dollar}u{dollar} of the image. Then, we cluster the cartoon {dollar}u{dollar} into different parts to obtain the final contour of different phases of the image. Moreover, {dollar}u{dollar} from the first stage is unique because of the convexity of the new model, and it needs to be computed only once whenever the thresholds and the number of the phases {dollar}K{dollar} in the second stage change. We implement the simulation on the synthetic and real images to show that our model outperforms some existed segmentation models in both precision and computational time
13
Schofield, Andrew John. "Neural network models for texture segmentation and target detection." Thesis, Keele University, 1993. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.358048.
Повний текст джерела
14
Subakan, Ozlem N. "Continuous mixture models for feature preserving smoothing and segmentation." [Gainesville, Fla.] : University of Florida, 2009. http://purl.fcla.edu/fcla/etd/UFE0024915.
Повний текст джерела
15
LIMA, MAXIMILIANO MORENO. "FUZZY MODELS IN SEGMENTATION AND ANALYSIS OF BANK MARKETING." PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO, 2008. http://www.maxwell.vrac.puc-rio.br/Busca_etds.php?strSecao=resultado&nrSeq=12290@1.
Повний текст джерелаАнотація:
COORDENAÇÃO DE APERFEIÇOAMENTO DO PESSOAL DE ENSINO SUPERIOREste trabalho tem como principal objetivo propor e desenvolver uma metodologia baseada em modelos fuzzy para a segmentação e caracterização dos segmentos que compõem o mercado bancário, permitindo um amplo conhecimento dos perfis de clientes, melhor adaptação das ofertas ao mercado e, conseqüentemente, melhores retornos financeiros. A metodologia proposta nesta dissertação pode ser dividida em três módulos principais: coleta e tratamento dos dados; definição dos segmentos; e caracterização e classificação dos segmentos. O primeiro módulo, denominado coleta e tratamento dos dados, abrange as pesquisas de marketing utilizadas na coleta dos dados e a aplicação de técnicas de pré-processamento de dados, para a limpeza (remoção de outliers e missing values) e normalização dos dados. O módulo de definição dos segmentos emprega o modelo fuzzy de agrupamento Fuzzy C-Means (FCM) na descoberta de grupos de clientes que apresentem características semelhantes. A escolha deste modelo de agrupamento deve-se à possibilidade de análise dos graus de pertinência de cada cliente em relação aos diferentes grupos, identificando os clientes entre segmentos e, conseqüentemente, elaborando ações efetivas para a sua transição ou manutenção nos segmentos de interesse. O módulo de caracterização e classificação dos segmentos é baseado em um Sistema de Inferência Fuzzy. Na primeira etapa deste módulo são selecionadas as variáveis mais relevantes, do ponto de vista da informação, para sua aplicação no processo de extração de regras. As regras extraídas para a caracterização dos segmentos são posteriormente utilizadas na construção de um sistema de inferência fuzzy dedicado à classificação de novos clientes. Este sistema permite que os analistas de marketing contribuam com novas regras ou modifiquem as já extraídas, tornando o modelo mais robusto e a segmentação de mercado uma ferramenta acessível a todos que dela se servem. A metodologia foi aplicada na segmentação de mercado do Banco da Amazônia, um banco estatal que atua na Amazônia Legal, cujo foco prioritário constitui o fomento da região. Avaliando a aplicação dos modelos fuzzy no estudo de caso, observam-se bons resultados na definição dos segmentos, com médias de valor de silhueta de 0,7, e na classificação da base de clientes, com acurácia de 100%. Adicionalmente, o uso destes modelos na segmentação de mercado possibilitou a análise dos clientes que estão entre segmentos e a caracterização desses segmentos por meio de uma base de regras, ampliando as análises dos analistas de marketing.
The main aim of this work is to propose and develop a methodology base don fuzzy models for segmentation and characterization of segments comprising the bank segment, allowing broad knowledge of client profiles, better suiting market needs, hence offering better financial results. The methodology proposed in this work may be divided into three main modules: data collection and treatment; definition of segments; and characterization and classification of segments. The first module, denominated data collection and treatment, encompasses marketing research used in data collection and application of techniques for pre-processing of data, for data trimming (removal of outliers and missing values) and normalization. The definition of segments adopts the Fuzzy C-Means (FCM) grouping model in identifying groups of clients with similar characteristics. The choice for this grouping model is due to the possibility of analyzing the membership coefficient of each client in connection with the different groups, thus identifying clients among segments and consequently elaborating effective actions for their transition to or maintenance in the segments of interest. The module of characterization and classification of segments is based on a Fuzzy Inference System. In the first stage, the most relevant variables from the information standpoint are selected, for application in the process of rule extraction. The rules extracted are then used in the construction of a fuzzy inference system dedicated to classifying new clients. This system allows marketing analysts to contribute with new rules or modify those already extracted, making the model more robust and the turning market segmentation into a tool accessible to all using it. This methodology was applied in the market segmentation of Banco da Amazônia, stte- contrlled bank acting in the Amazon region, with main focus of which is fostering the region´s development. The application of fuzzy models in the case study generated good results in the definition of segments, with average silhouette value of 0.7, and accuracy of 100% for client base classification. Furthermore, the use of these models in market segmentation allowed the analysis of clients classified between segments and the characterization of those segments by means of a set of rules, improving the analyses made by marketing analysts.
16
Budvytis, Ignas. "Novel probabilistic graphical models for semi-supervised video segmentation." Thesis, University of Cambridge, 2013. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.648293.
Повний текст джерела
17
Bou, Albert. "Deep Learning models for semantic segmentation of mammography screenings." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-265652.
Повний текст джерелаАнотація:
This work explores the performance of state-of-the-art semantic segmentation models on mammographic imagery. It does so by comparing several reference semantic segmentation deep learning models on a newly proposed medical dataset of mammograpgy screenings. All models are re-implemented in Tensorflow and validated first on the benchmark dataset Cityscapes. The new medical image corpus was gathered and annotated at the Science for Life Laboratory in Stockholm. In addition, this master thesis shows that it is possible to boost segmentation performance by training the models in an adversarial manner after reaching convergence in the classical training framework.Denna uppsats undersöker hur väl moderna metoder presterar på semantisk segmentering av mammografibilder. Detta görs genom att utvärdera flera semantiska segmenteringsmetoder på ett dataset som är framtaget under detta examensarbete. Utvärderingarna genomförs genom att återimplementera flertalet semantiska segmenteringsmodeller för djupinlärning i Tensorflow och algoritmerna valideras på referensdatasetet Cityscapes. Därefter tränas modellerna också på det dataset med medicinska mammografi-bilder som är samlat och annoterat vid Science for Life Laboratory i Stockholm. Dessutom visar detta examensarbete att det är möjligt att öka segmenteringsprestandan genom att använda en adversarial träningsmetod efter att den klassiska träningsalgoritmen har konvergerat.
18
Rada, Lavdie. "Variational models and numerical algorithms for selective image segmentation." Thesis, University of Liverpool, 2013. http://livrepository.liverpool.ac.uk/11093/.
Повний текст джерелаАнотація:
This thesis deals with the numerical solution of nonlinear partial differential equations and their application in image processing. The differential equations we deal with here arise from the minimization of variational models for image restoration techniques (such as denoising) and recognition of objects techniques (such as segmentation). Image denoising is a technique aimed at restoring a digital image that has been contaminated by noise while segmentation is a fundamental task in image analysis responsible for partitioning an image as sub-regions or representing the image into something that is more meaningful and easier to analyze such as extracting one or more specific objects of interest in images based on relevant information or a desired feature. Although there has been a lot of research in the restoration of images, the performance of such methods is still poor, especially when the images have a high level of noise or when the algorithms are slow. Task of the segmentation is even more challenging problem due to the difficulty of delineating, even manually, the contours of the objects of interest. The problems are often due to low contrast, fuzzy contours, similar intensities with adjacent objects, or the objects to be extracted having no real contours. The first objective of this work is to develop fast image restoration and segmentation methods which provide better denoising and fast and robust performance for image segmentation. The contribution presented here is the development of a restarted homotopy analysis method which has been designed to be easily adaptable to various types of image processing problems. As a second research objective we propose a framework for image selective segmentation which partitions an image based on the information known in advance of the object/objects to be extracted (for example the left kidney is the target to be extracted in a CT image and the prior knowledge is a few markers in this object of interest). This kind of segmentation appears especially in medical applications. Medical experts usually estimate and manually draw the boundaries of the organ/organs based on their experience. Our aim is to introduce automatic segmentation of the object of interest as a contribution not only to the way doctors and surgeons diagnose and operate but to other fields as well. The proposed methods showed success in segmenting different objects and perform well in different types of images not only in two-dimensional but in three-dimensional images as well.
19
Badshah, Noor. "Fast iterative methods for variational models in image segmentation." Thesis, University of Liverpool, 2009. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.501735.
Повний текст джерелаАнотація:
Image segmentation is an important branch of computer vision. It aims at extracting meaningM objects lying in images either by dividing images 5 into contiguous semantic regions, or by extracting one or more specific objects in images such as left kidney in CT image. The image segmentation task is, in general, very difficult to achieve since natural images are diverse and complex, and the way we perceive them varies according to individuals.
20
Pereañez, Marco. "Enlargement, subdivision and individualization of statistical shape models: Application to 3D medical image segmentation." Doctoral thesis, Universitat Pompeu Fabra, 2017. http://hdl.handle.net/10803/441754.
Повний текст джерелаАнотація:
This thesis presents three original and complementary approaches to enhance the quality
of Statistical Shape Models (SSMs), that improve the accuracy of medical image
segmentation in challenging applications. First, we enhance the statistical richness of SSMs
by developing a technique capable of merging the shape representations and statistical
properties of several pre-existing models with no original or additional raw data. Second,
we enhance the geometrical quality of SSMs by developing a framework for modeling
simultaneously both global and local characteristics of highly complex and/or multi-part
anatomical shapes. Last, we improve the specificity of SSMs for specific subjects by
integrating individual-specific non-imaging metadata such as demographic, clinical and
behavioral variables into the SSM construction and image segmentation tasks. These
techniques are demonstrated and validated by considering various imaging modalities such
as magnetic resonance imaging (MRI) and computed tomography (CT), and different
complex anatomies, including the human heart, brain and spine.Esta tesis presenta tres propuestas originales y complementarias para mejorar la calidad de los modelos estadísticos de formas (SSMs) que mejoran la precisión de la segmentación de la imagen médica en aplicaciones difíciles. Proponemos, primero, mejorar la riqueza estadística de los SSMs por medio de una técnica para unir la representación de forma y las propiedades estadísticas de muchos modelos pre-existentes sin observaciones adicionales. Segundo, mejorar la representacion geométrica de los SSMs modelando simultáneamente las características globales y locales del objecto o de multiples anatomias. Por último, mejorar la especificidad de los SSMs mediante la integración de metadatos del paciente no derivados de la imagen, tales como, variables demográficas, conductuales y de entorno clínico, en la construcción de los modelos. Estas técnicas son demostradas y validadas en imágenes de resonancia magnética (MRI) y tomografía computarizada (CT) y en anatomias como el corazón, el cerebro y la espina dorsal humanos.
21
Wang, Chaohui. "Distributed and Higher-Order Graphical Models : towards Segmentation, Tracking, Matching and 3D Model Inference." Phd thesis, Ecole Centrale Paris, 2011. http://tel.archives-ouvertes.fr/tel-00658765.
Повний текст джерелаАнотація:
This thesis is devoted to the development of graph-based methods that address several of the most fundamental computer vision problems, such as segmentation, tracking, shape matching and 3D model inference. The first contribution of this thesis is a unified, single-shot optimization framework for simultaneous segmentation, depth ordering and multi-object tracking from monocular video sequences using a pairwise Markov Random Field (MRF). This is achieved through a novel 2.5D layered model where object-level and pixel-level representations are seamlessly combined through local constraints. Towards introducing high-level knowledge, such as shape priors, we then studied the problem of non-rigid 3D surface matching. The second contribution of this thesis consists of a higher-order graph matching formulation that encodes various measurements of geometric/appearance similarities and intrinsic deformation errors. As the third contribution of this thesis, higher-order interactions were further considered to build pose-invariant statistical shape priors and were exploited for the development of a novel approach for knowledge-based 3D segmentation in medical imaging which is invariant to the global pose and the initialization of the shape model. The last contribution of this thesis aimed to partially address the influence of camera pose in visual perception. To this end, we introduced a unified paradigm for 3D landmark model inference from monocular 2D images to simultaneously determine both the optimal 3D model and the corresponding 2D projections without explicit estimation of the camera viewpoint, which is also able to deal with misdetections/occlusions
22
Kozinski, Mateusz. "Segmentation of facade images with shape priors." Thesis, Paris Est, 2015. http://www.theses.fr/2015PESC1017/document.
Повний текст джерелаАнотація:
L'objectif de cette thèse concerne l'analyse automatique d'images de façades de bâtiments à partir de descriptions formelles à priori de formes géométriques. Ces informations suggérées par un utilisateur permettent de modéliser, de manière formelle, des contraintes spatiales plus ou moins dures quant à la segmentation sémantique produite par le système. Ceci permet de se défaire de deux principaux écueils inhérents aux méthodes d'analyse de façades existantes qui concernent d'une part la coûteuse fidélité de la segmentation résultante aux données visuelles de départ, d'autre part, la spécificité architecturale des règles imposées lors du processus de traitement. Nous proposons d'explorer au travers de cette thèse, différentes méthodes alternatives à celles proposées dans la littérature en exploitant un formalisme de représentation d'à priori de haut niveau d'abstraction, les propriétés engendrées par ces nouvelles méthodes ainsi que les outils de résolution mis en œuvres par celles-ci. Le système résultant est évalué tant quantitativement que qualitativement sur de multiples bases de données standards et par le biais d'études comparatives à des approches à l'état de l'art en la matière. Parmi nos contributions, nous pouvons citer la combinaison du formalisme des grammaires de graphes exprimant les variations architecturales de façades de bâtiments et les modèles graphiques probabilistes modélisant l'énergie attribuée à une configuration paramétrique donnée, dans un schéma d'optimisation par minimisation d'énergie; ainsi qu'une nouvelle approche par programmation linéaire d'analyse avec à priori de formes. Enfin, nous proposons un formalisme flexible de ces à priori devançant de par ses performances les méthodes à l'état de l'art tout en combinant les avantages de la généricité de contraintes simples manuellement imposées par un utilisateur, à celles de la précision de la segmentation finale qui se faisait jusqu'alors au prix d'un encodage préliminaire restrictif de règles grammaticales complexes propres à une famille architecturale donnée. Le système décrit permet également de traiter avec robustesse des scènes comprenant des objets occultants et pourrait encore être étendu notamment afin de traiter l'extension tri-dimensionnelle de la sémantisation d'environnements urbains sous forme de nuages de points 3D ou d'une analyse multi-image de bâtimentsThe aim of this work is to propose a framework for facade segmentation with user-defined shape priors. In such a framework, the user specifies a shape prior using a rigorously defined shape prior formalism. The prior expresses a number of hard constraints and soft preference on spatial configuration of segments, constituting the final segmentation. Existing approaches to the problem are affected by a compromise between the type of constraints, the satisfaction of which can be guaranteed by the segmentation algorithm, and the capability to approximate optimal segmentations consistent with a prior. In this thesis we explore a number of approaches to facade parsing that combine prior formalism featuring high expressive power, guarantees of conformance of the resulting segmentations to the prior, and effective inference. We evaluate the proposed algorithms on a number of datasets. Since one of our focus points is the accuracy gain resulting from more effective inference algorithms, we perform a fair comparison to existing methods, using the same data term. Our contributions include a combination of graph grammars for expressing variation of facade structure with graphical models encoding the energy of models of given structures for different positions of facade elements. We also present the first linear formulation of facade parsing with shape priors. Finally, we propose a shape prior formalism that enables formulating the problem of optimal segmentation as the inference in a Markov random field over the standard four-connected grid of pixels. The last method advances the state of the art by combining the flexibility of a user-defined grammar with segmentation accuracy that was reserved for frameworks with pre-defined priors before. It also enables handling occlusions by simultaneously recovering the structure of the occluded facade and segmenting the occluding objects. We believe that it can be extended in many directions, including semantizing three-dimensional point clouds and parsing images of general urban scenes
23
Demirkol, Onur Ali. "Segmentation Of Torso Ct Images." Master's thesis, METU, 2006. http://etd.lib.metu.edu.tr/upload/2/12607431/index.pdf.
Повний текст джерелаАнотація:
Medical imaging modalities provide effective information for anatomic or
metabolic activity of tissues and organs in the body. Therefore, medical
imaging technology is a critical component in diagnosis and treatment of
various illnesses. Medical image segmentation plays an important role in
converting medical images into anatomically, functionally or surgically
identifiable structures, and is used in various applications. In this study,
some of the major medical image segmentation methods are examined and
applied to 2D CT images of upper torso for segmentation of heart, lungs,
bones, and muscle and fat tissues. The implemented medical image
segmentation methods are thresholding, region growing, watershed
transformation, deformable models and a hybrid methodwatershed transformation and region merging. Moreover, a comparative analysis is performed among these methods to obtain the most efficient segmentation method for each tissue and organ in torso. Some improvements are proposed for increasing accuracy of some image segmentation methods.
24
Soliman, Ahmed Talaat Elsayed. "Hidden Markov Models Based Segmentation of Brain Magnetic Resonance Imaging." Scholarly Repository, 2007. http://scholarlyrepository.miami.edu/oa_theses/80.
Повний текст джерелаАнотація:
Two brain segmentation approaches based on Hidden Markov Models are proposed. The first approach aims to segment normal brain 3D multi-channel MR images into three tissues WM, GM, and CSF. Linear Discriminant Analysis, LDA, is applied to separate voxels belonging to different tissues as well as to reduce their features vector size. The second approach aims to detect MS lesions in Brain 3D multi-channel MR images and to label WM, GM, and CSF tissues. Preprocessing is applied in both approaches to reduce the noise level and to address sudden intensity and global intensity correction. The proposed techniques are tested using 3D images from Montereal BrainWeb data set. In the first approach, the results were numerically assessed and compared to results reported using techniques based on single channel data and applied to the same data sets. The results obtained using the multi channel HMM-based algorithm were better than the results reported for single channel data in terms of an objective measure of overlap, Dice coefficient, compared to other methods. In the second approach, the segmentation accuracy is measured using Dice coefficient and total lesions load percentage
25
Gomes, Vicente S. A. "Global optimisation techniques for image segmentation with higher order models." Thesis, University College London (University of London), 2011. http://discovery.ucl.ac.uk/1334450/.
Повний текст джерелаАнотація:
Energy minimisation methods are one of the most successful approaches to image segmentation. Typically used energy functions are limited to pairwise interactions due to the increased complexity when working with higher-order functions. However, some important assumptions about objects are not translatable to pairwise interactions. The goal of this thesis is to explore higher order models for segmentation that are applicable to a wide range of objects. We consider: (1) a connectivity constraint, (2) a joint model over the segmentation and the appearance, and (3) a model for segmenting the same object in multiple images. We start by investigating a connectivity prior, which is a natural assumption about objects. We show how this prior can be formulated in the energy minimisation framework and explore the complexity of the underlying optimisation problem, introducing two different algorithms for optimisation. This connectivity prior is useful to overcome the “shrinking bias” of the pairwise model, in particular in interactive segmentation systems. Secondly, we consider an existing model that treats the appearance of the image segments as variables. We show how to globally optimise this model using a Dual Decomposition technique and show that this optimisation method outperforms existing ones. Finally, we explore the current limits of the energy minimisation framework. We consider the cosegmentation task and show that a preference for object-like segmentations is an important addition to cosegmentation. This preference is, however, not easily encoded in the energy minimisation framework. Instead, we use a practical proposal generation approach that allows not only the inclusion of a preference for object-like segmentations, but also to learn the similarity measure needed to define the cosegmentation task. We conclude that higher order models are useful for different object segmentation tasks. We show how some of these models can be formulated in the energy minimisation framework. Furthermore, we introduce global optimisation methods for these energies and make extensive use of the Dual Decomposition optimisation approach that proves to be suitable for this type of models.
26
Corneli, Marco. "Dynamic stochastic block models, clustering and segmentation in dynamic graphs." Thesis, Paris 1, 2017. http://www.theses.fr/2017PA01E012/document.
Повний текст джерелаАнотація:
Cette thèse porte sur l’analyse de graphes dynamiques, définis en temps discret ou continu. Nous introduisons une nouvelle extension dynamique du modèle a blocs stochastiques (SBM), appelée dSBM, qui utilise des processus de Poisson non homogènes pour modéliser les interactions parmi les paires de nœuds d’un graphe dynamique. Les fonctions d’intensité des processus ne dépendent que des classes des nœuds comme dans SBM. De plus, ces fonctions d’intensité ont des propriétés de régularité sur des intervalles temporels qui sont à estimer, et à l’intérieur desquels les processus de Poisson redeviennent homogènes. Un récent algorithme d’estimation pour SBM, qui repose sur la maximisation d’un critère exact (ICL exacte) est ici adopté pour estimer les paramètres de dSBM et sélectionner simultanément le modèle optimal. Ensuite, un algorithme exact pour la détection de rupture dans les séries temporelles, la méthode «pruned exact linear time» (PELT), est étendu pour faire de la détection de rupture dans des données de graphe dynamique selon le modèle dSBM. Enfin, le modèle dSBM est étendu ultérieurement pour faire de l’analyse de réseau textuel dynamique. Les réseaux sociaux sont un exemple de réseaux textuels: les acteurs s’échangent des documents (posts, tweets, etc.) dont le contenu textuel peut être utilisé pour faire de la classification et détecter la structure temporelle du graphe dynamique. Le modèle que nous introduisons est appelé «dynamic stochastic topic block model» (dSTBM)This thesis focuses on the statistical analysis of dynamic graphs, both defined in discrete or continuous time. We introduce a new extension of the stochastic block model (SBM) for dynamic graphs. The proposed approach, called dSBM, adopts non homogeneous Poisson processes to model the interaction times between pairs of nodes in dynamic graphs, either in discrete or continuous time. The intensity functions of the processes only depend on the node clusters, in a block modelling perspective. Moreover, all the intensity functions share some regularity properties on hidden time intervals that need to be estimated. A recent estimation algorithm for SBM, based on the greedy maximization of an exact criterion (exact ICL) is adopted for inference and model selection in dSBM. Moreover, an exact algorithm for change point detection in time series, the "pruned exact linear time" (PELT) method is extended to deal with dynamic graph data modelled via dSBM. The approach we propose can be used for change point analysis in graph data. Finally, a further extension of dSBM is developed to analyse dynamic net- works with textual edges (like social networks, for instance). In this context, the graph edges are associated with documents exchanged between the corresponding vertices. The textual content of the documents can provide additional information about the dynamic graph topological structure. The new model we propose is called "dynamic stochastic topic block model" (dSTBM).Graphs are mathematical structures very suitable to model interactions between objects or actors of interest. Several real networks such as communication networks, financial transaction networks, mobile telephone networks and social networks (Facebook, Linkedin, etc.) can be modelled via graphs. When observing a network, the time variable comes into play in two different ways: we can study the time dates at which the interactions occur and/or the interaction time spans. This thesis only focuses on the first time dimension and each interaction is assumed to be instantaneous, for simplicity. Hence, the network evolution is given by the interaction time dates only. In this framework, graphs can be used in two different ways to model networks. Discrete time […] Continuous time […]. In this thesis both these perspectives are adopted, alternatively. We consider new unsupervised methods to cluster the vertices of a graph into groups of homogeneous connection profiles. In this manuscript, the node groups are assumed to be time invariant to avoid possible identifiability issues. Moreover, the approaches that we propose aim to detect structural changes in the way the node clusters interact with each other. The building block of this thesis is the stochastic block model (SBM), a probabilistic approach initially used in social sciences. The standard SBM assumes that the nodes of a graph belong to hidden (disjoint) clusters and that the probability of observing an edge between two nodes only depends on their clusters. Since no further assumption is made on the connection probabilities, SBM is a very flexible model able to detect different network topologies (hubs, stars, communities, etc.)
27
Ivins, James P. "Statistical snakes: active region models." Thesis, University of Sheffield, 1996. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.484310.
Повний текст джерела
28
Buchta, Christian, and Sara Dolnicar. "Learning by simulation. Computer simulations for strategic marketing decision support in tourism." SFB Adaptive Information Systems and Modelling in Economics and Management Science, WU Vienna University of Economics and Business, 2003. http://epub.wu.ac.at/1718/1/document.pdf.
Повний текст джерелаАнотація:
This paper describes the use of corporate decision and strategy simulations as a decision-support instrument under varying market conditions in the tourism industry. It goes on to illustrate this use of simulations with an experiment which investigates how successful different market segmentation approaches are in destination management. The experiment assumes a competitive environment and various cycle-length conditions with regard to budget and strategic planning. Computer simulations prove to be a useful management tool, allowing customized experiments which provide insight into the functioning of the market and therefore represent an interesting tool for managerial decision support. The main drawback is the initial setup of a customized computer simulation, which is time-consuming and involves defining parameters with great care in order to represent the actual market environment and to avoid excessive complexity in testing cause-effect-relationships. (author's abstract)Series: Report Series SFB "Adaptive Information Systems and Modelling in Economics and Management Science"
29
Molin, Joel. "Foreground Segmentation of Moving Objects." Thesis, Linköping University, Department of Electrical Engineering, 2010. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-52544.
Повний текст джерелаАнотація:
Foreground segmentation is a common first step in tracking and surveillance applications. The purpose of foreground segmentation is to provide later stages of image processing with an indication of where interesting data can be found. This thesis is an investigation of how foreground segmentation can be performed in two contexts: as a pre-step to trajectory tracking and as a pre-step in indoor surveillance applications.
Three methods are selected and detailed: a single Gaussian method, a Gaussian mixture model method, and a codebook method. Experiments are then performed on typical input video using the methods. It is concluded that the Gaussian mixture model produces the output which yields the best trajectories when used as input to the trajectory tracker. An extension is proposed to the Gaussian mixture model which reduces shadow, improving the performance of foreground segmentation in the surveillance context.
30
Patenaude, Brian Matthew. "Bayesian statistical models of shape and appearance for subcortical brain segmentation." Thesis, University of Oxford, 2007. http://ora.ox.ac.uk/objects/uuid:52f5fee0-60e8-4387-9560-728843e187b3.
Повний текст джерелаАнотація:
Our motivation is to develop an automated technique for the segmentation of sub-cortical human brain structures from MR images. To this purpose, models of shape-and-appearance are constructed and fit to new image data. The statistical models are trained from 317 manually labelled T1-weighted MR images. Shape is modelled using a surface-based point distribution model (PDM) such that the shape space is constrained to the linear combination of the mean shape and eigenvectors of the vertex coordinates. In addition, to model intensity at the structural boundary, intensities are sampled along the surface normal from the underlying image. We propose a novel Bayesian appearance model whereby the relationship between shape and intensity are modelled via the conditional distribution of intensity given shape. Our fully probabilistic approach eliminates the need for arbitrary weightings between shape and intensity as well as for tuning parameters that specify the relative contribution between the use of shape constraints and intensity information. Leave-one-out cross-validation is used to validate the model and fitting for 17 structures. The PDM for shape requires surface parameterizations of the volumetric, manual labels such that vertices retain a one-to-one correspondence across the training subjects. Surface parameterizations with correspondence are generated through the use of deformable models under constraints that embed the correspondence criterion within the deformation process. A novel force that favours equal-area triangles throughout the mesh is introduced. The force adds stability to the mesh such that minimal smoothing or within-surface motion is required. The use of the PDM for segmentation across a series of subjects results in a set surfaces that retain point correspondence. The correspondence facilitates landmark-based shape analysis. Amongst other metrics, vertex-wise multivariate statistics and discriminant analysis are used to investigate local and global size and shape differences between groups. The model is fit, and shape analysis is applied to two clinical datasets.
31
Kantedal, Simon. "Evaluating Segmentation of MR Volumes Using Predictive Models and Machine Learning." Thesis, Linköpings universitet, Institutionen för medicinsk teknik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-171102.
Повний текст джерелаАнотація:
A reliable evaluation system is essential for every automatic process. While techniques for automatic segmentation of images have been extensively researched in recent years, evaluation of the same has not received an equal amount of attention. Amra Medical AB has developed a system for automatic segmentation of magnetic resonance (MR) images of human bodies using an atlas-based approach. Through their software, Amra is able to derive body composition measurements, such as muscle and fat volumes, from the segmented MR images. As of now, the automatic segmentations are quality controlled by clinical experts to ensure their correctness. This thesis investigates the possibilities to leverage predictive modelling to reduce the need for a manual quality control (QC) step in an otherwise automatic process. Two different regression approaches have been implemented as a part of this study: body composition measurement prediction (BCMP) and manual correction prediction (MCP). BCMP aims at predicting the derived body composition measurements and comparing the predictions to actual measurements. The theory is that large deviations between the predictions and the measurements signify an erroneously segmented sample. MCP instead tries to directly predict the amount of manual correction needed for each sample. Several regression models have been implemented and evaluated for the two approaches. Comparison of the regression models shows that local linear regression (LLR) is the most performant model for both BCMP and MCP. The results show that the inaccuracies in the BCMP-models, in practice, renders this approach useless. MCP proved to be a far more viable approach; using MCP together with LLR achieves a high true positive rate with a reasonably low false positive rate for several body composition measurements. These results suggest that the type of system developed in this thesis has the potential to reduce the need for manual inspections of the automatic segmentation masks.
32
Zeng, Jingying. "Latent Factor Models for Recommender Systems and Market Segmentation Through Clustering." The Ohio State University, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=osu1491255524283942.
Повний текст джерела
33
Axberg, Elin, and Ida Klerstad. "Similarity models for atlas-based segmentation of whole-body MRI volumes." Thesis, Linköpings universitet, Institutionen för medicinsk teknik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-172792.
Повний текст джерелаАнотація:
In order to analyse body composition of MRI (Magnetic Resonance Imaging) volumes, atlas-based segmentation is often used to retrieve information from specific organs or anatomical regions. The method behind this technique is to use an already segmented image volume, an atlas, to segment a target image volume by registering the volumes to each other. During this registration a deformation field will be calculated, which is applied to a segmented part of the atlas, resulting in the same anatomical segmentation in the target. The drawback with this method is that the quality of the segmentation is highly dependent on the similarity between the target and the atlas, which means that many atlases are needed to obtain good segmentation results in large sets of MRI volumes. One potential solution to overcome this problem is to create the deformation field between a target and an atlas as a sequence of small deformations between more similar bodies. In this master thesis a new method for atlas-based segmentation has been developed, with the anticipation of obtaining good segmentation results regardless of the level of similarity between the target and the atlas. In order to do so, 4000 MRI volumes were used to create a manifold of human bodies, which represented a large variety of different body types. These MRI volumes were compared to each other and the calculated similarities were saved in matrices called similarity models. Three different similarity measures were used to create the models which resulted in three different versions of the model. In order to test the hypothesis of achieving good segmentation results when the deformation field was constructed as a sequence of small deformations, the similarity models were used to find the shortest path (the path with the least dissimilarity) between a target and an atlas in the manifold. In order to evaluate the constructed similarity models, three MRI volumes were chosen as atlases and 100 MRI volumes were randomly picked to be used as targets. The shortest paths between these volumes were used to create the deformation fields as a sequence of small deformations. The created fields were then used to segment the anatomical regions ASAT (abdominal subcutaneous adipose tissue), LPT (left posterior thigh) and VAT (visceral adipose tissue). The segmentation performance was measured with Dice Index, where segmentations constructed at AMRA Medical AB were used as ground truth. In order to put the results in relation to another segmentation method, direct deformation fields between the targets and the atlases were also created and the segmentation results were compared to the ground truth with the Dice Index. Two different types of transformation methods, one non-parametric and one affine transformation, were used to create the deformation fields in this master thesis. The evaluation showed that good segmentation results can be achieved for the segmentation of VAT for one of the constructed similarity models. These results were obtained when a non-parametric registration method was used to create the deformation fields. In order to achieve similar results for an affine registration and to improve the segmentation of other anatomical regions, further investigations are needed.
34
Lind, Johan. "Make it Meaningful : Semantic Segmentation of Three-Dimensional Urban Scene Models." Thesis, Linköpings universitet, Datorseende, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-143599.
Повний текст джерелаАнотація:
Semantic segmentation of a scene aims to give meaning to the scene by dividing it into meaningful — semantic — parts. Understanding the scene is of great interest for all kinds of autonomous systems, but manual annotation is simply too time consuming, which is why there is a need for an alternative approach. This thesis investigates the possibility of automatically segmenting 3D-models of urban scenes, such as buildings, into a predetermined set of labels. The approach was to first acquire ground truth data by manually annotating five 3D-models of different urban scenes. The next step was to extract features from the 3D-models and evaluate which ones constitutes a suitable feature space. Finally, three supervised learners were implemented and evaluated: k-Nearest Neighbour (KNN), Support Vector Machine (SVM) and Random Classification Forest (RCF). The classifications were done point-wise, classifying each 3D-point in the dense point cloud belonging to the model being classified. The result showed that the best suitable feature space is not necessarily the one containing all features. The KNN classifier got the highest average accuracy overall models — classifying 42.5% of the 3D points correct. The RCF classifier managed to classify 66.7% points correct in one of the models, but had worse performance for the rest of the models and thus resulting in a lower average accuracy compared to KNN. In general, KNN, SVM, and RCF seemed to have different benefits and drawbacks. KNN is simple and intuitive but by far the slowest classifier when dealing with a large set of training data. SVM and RCF are both fast but difficult to tune as there are more parameters to adjust. Whether the reason for obtaining the relatively low highest accuracy was due to the lack of ground truth training data, unbalanced validation models, or the capacity of the learners, was never investigated due to a limited time span. However, this ought to be investigated in future studies.
35
Ruckert, Daniel. "Segmentation and tracking in cardiovascular images using geometrically deformable models and templates." Thesis, Imperial College London, 1998. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.286210.
Повний текст джерела
36
Trajkovska, Vera [Verfasser], and Christoph [Akademischer Betreuer] Schnörr. "Learning Probabilistic Graphical Models for Image Segmentation / Vera Trajkovska ; Betreuer: Christoph Schnörr." Heidelberg : Universitätsbibliothek Heidelberg, 2017. http://d-nb.info/1177689987/34.
Повний текст джерела
37
Rathke, Fabian [Verfasser], and Christoph [Akademischer Betreuer] Schnörr. "Probabilistic Graphical Models for Medical Image Segmentation / Fabian Rathke ; Betreuer: Christoph Schnörr." Heidelberg : Universitätsbibliothek Heidelberg, 2015. http://d-nb.info/1180395042/34.
Повний текст джерела
38
Ferrand, M. "Data-driven, memory-based computational models of human segmentation of musical melody." Thesis, University of Edinburgh, 2006. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.650869.
Повний текст джерелаАнотація:
This study seeks the development of a system capable of performing melodic segmentation in an unsupervised way, by learning from non-annotated musical data. Probabilistic learning methods have been widely used to acquire regularities in large sets of data, with many successful applications in language and speech processing. Some of these applications have found their counterparts in music research and have been used for music prediction and generation, music retrieval or music analysis, but seldom to model perceptual and cognitive aspects of music listening. We present some preliminary experiments on melodic segmentation, which highlight the importance of memory and the role of learning in music listening. These experiments have motivated the development of a computational model for melodic segmentation based on a probabilistic learning paradigm. This model uses a Mixed-memory Markov Model to estimate sequence probabilities from pitch and time-based parametric descriptions of melodic data. We follow the assumption that listeners’ perception of feature salience in melodies is strongly related to expectation. Moreover, we conjecture that outstanding entropy variations of certain melodic features coincide with segmentation boundaries as indicated by listeners. Model segmentation predictions are compared with results of a listening study on melodic segmentation carried out with real listeners. Overall results show that changes in prediction entropy along the pieces exhibit significant correlation with the listeners’ segmentation boundaries. Although the model relies only on information theoretic principles to make predictions on the location of segmentation boundaries, it was found that most predicted segments could be matched with boundaries of groupings usually attributed to Gestalt rules. These results question previous research supporting a separation between learning-based and innate bottom-up processes of melodic grouping, and suggesting that some of these latter processes can emerge from acquired regularities in melodic data.
39
Boussaid, Haithem. "Efficient inference and learning in graphical models for multi-organ shape segmentation." Thesis, Châtenay-Malabry, Ecole centrale de Paris, 2015. http://www.theses.fr/2015ECAP0002/document.
Повний текст джерелаАнотація:
Cette thèse explore l’utilisation des modèles de contours déformables pour la segmentation basée sur la forme des images médicales. Nous apportons des contributions sur deux fronts: dans le problème de l’apprentissage statistique, où le modèle est formé à partir d’un ensemble d’images annotées, et le problème de l’inférence, dont le but est de segmenter une image étant donnée un modèle. Nous démontrons le mérite de nos techniques sur une grande base d’images à rayons X, où nous obtenons des améliorations systématiques et des accélérations par rapport à la méthode de l’état de l’art. Concernant l’apprentissage, nous formulons la formation de la fonction de score des modèles de contours déformables en un problème de prédiction structurée à grande marge et construisons une fonction d’apprentissage qui vise à donner le plus haut score à la configuration vérité-terrain. Nous intégrons une fonction de perte adaptée à la prédiction structurée pour les modèles de contours déformables. En particulier, nous considérons l’apprentissage avec la mesure de performance consistant en la distance moyenne entre contours, comme une fonction de perte. L’utilisation de cette fonction de perte au cours de l’apprentissage revient à classer chaque contour candidat selon sa distance moyenne du contour vérité-terrain. Notre apprentissage des modèles de contours déformables en utilisant la prédiction structurée avec la fonction zéro-un de perte surpasse la méthode [Seghers et al. 2007] de référence sur la base d’images médicales considérée [Shiraishi et al. 2000, van Ginneken et al. 2006]. Nous démontrons que l’apprentissage avec la fonction de perte de distance moyenne entre contours améliore encore plus les résultats produits avec l’apprentissage utilisant la fonction zéro-un de perte et ce d’une quantité statistiquement significative.Concernant l’inférence, nous proposons des solveurs efficaces et adaptés aux problèmes combinatoires à variables spatiales discrétisées. Nos contributions sont triples: d’abord, nous considérons le problème d’inférence pour des modèles graphiques qui contiennent des boucles, ne faisant aucune hypothèse sur la topologie du graphe sous-jacent. Nous utilisons un algorithme de décomposition-coordination efficace pour résoudre le problème d’optimisation résultant: nous décomposons le graphe du modèle en un ensemble de sous-graphes en forme de chaines ouvertes. Nous employons la Méthode de direction alternée des multiplicateurs (ADMM) pour réparer les incohérences des solutions individuelles. Même si ADMM est une méthode d’inférence approximative, nous montrons empiriquement que notre implémentation fournit une solution exacte pour les exemples considérés. Deuxièmement, nous accélérons l’optimisation des modèles graphiques en forme de chaîne en utilisant l’algorithme de recherche hiérarchique A* [Felzenszwalb & Mcallester 2007] couplé avec les techniques d’élagage développés dans [Kokkinos 2011a]. Nous réalisons une accélération de 10 fois en moyenne par rapport à l’état de l’art qui est basé sur la programmation dynamique (DP) couplé avec les transformées de distances généralisées [Felzenszwalb & Huttenlocher 2004]. Troisièmement, nous intégrons A* dans le schéma d’ADMM pour garantir une optimisation efficace des sous-problèmes en forme de chaine. En outre, l’algorithme résultant est adapté pour résoudre les problèmes d’inférence augmentée par une fonction de perte qui se pose lors de l’apprentissage de prédiction des structure, et est donc utilisé lors de l’apprentissage et de l’inférence. [...]This thesis explores the use of discriminatively trained deformable contour models (DCMs) for shape-based segmentation in medical images. We make contributions in two fronts: in the learning problem, where the model is trained from a set of annotated images, and in the inference problem, whose aim is to segment an image given a model. We demonstrate the merit of our techniques in a large X-Ray image segmentation benchmark, where we obtain systematic improvements in accuracy and speedups over the current state-of-the-art. For learning, we formulate training the DCM scoring function as large-margin structured prediction and construct a training objective that aims at giving the highest score to the ground-truth contour configuration. We incorporate a loss function adapted to DCM-based structured prediction. In particular, we consider training with the Mean Contour Distance (MCD) performance measure. Using this loss function during training amounts to scoring each candidate contour according to its Mean Contour Distance to the ground truth configuration. Training DCMs using structured prediction with the standard zero-one loss already outperforms the current state-of-the-art method [Seghers et al. 2007] on the considered medical benchmark [Shiraishi et al. 2000, van Ginneken et al. 2006]. We demonstrate that training with the MCD structured loss further improves over the generic zero-one loss results by a statistically significant amount. For inference, we propose efficient solvers adapted to combinatorial problems with discretized spatial variables. Our contributions are three-fold:first, we consider inference for loopy graphical models, making no assumption about the underlying graph topology. We use an efficient decomposition-coordination algorithm to solve the resulting optimization problem: we decompose the model’s graph into a set of open, chain-structured graphs. We employ the Alternating Direction Method of Multipliers (ADMM) to fix the potential inconsistencies of the individual solutions. Even-though ADMMis an approximate inference scheme, we show empirically that our implementation delivers the exact solution for the considered examples. Second,we accelerate optimization of chain-structured graphical models by using the Hierarchical A∗ search algorithm of [Felzenszwalb & Mcallester 2007] couple dwith the pruning techniques developed in [Kokkinos 2011a]. We achieve a one order of magnitude speedup in average over the state-of-the-art technique based on Dynamic Programming (DP) coupled with Generalized DistanceTransforms (GDTs) [Felzenszwalb & Huttenlocher 2004]. Third, we incorporate the Hierarchical A∗ algorithm in the ADMM scheme to guarantee an efficient optimization of the underlying chain structured subproblems. The resulting algorithm is naturally adapted to solve the loss-augmented inference problem in structured prediction learning, and hence is used during training and inference. In Appendix A, we consider the case of 3D data and we develop an efficientmethod to find the mode of a 3D kernel density distribution. Our algorithm has guaranteed convergence to the global optimum, and scales logarithmically in the volume size by virtue of recursively subdividing the search space. We use this method to rapidly initialize 3D brain tumor segmentation where we demonstrate substantial acceleration with respect to a standard mean-shift implementation. In Appendix B, we describe in more details our extension of the Hierarchical A∗ search algorithm of [Felzenszwalb & Mcallester 2007] to inference on chain-structured graphs
40
Chen, Zhibin. "Segmentation of MRI images using non parametric deformable models integrating fuzzy technique." Reims, 2009. http://theses.univ-reims.fr/exl-doc/GED00001122.pdf.
Повний текст джерелаАнотація:
L'objectif de la thèse est de développer une méthode automatique pour segmenter les tissus cérébraux (la matière grise, la matière blanche et le liquide céphalo-rachidien) à partir des images IRM, fournissant ainsi des mesures quantitatives et précises du cerveau. Dans cette thèse, nous avons développé trois modèles déformables non-paramétriques en intégrant l'information statistique et l’information floue des images pour segmenter le cerveau en différents types de tissus. Nous présentons d'abord une méthode basée sur l’analyse de l'histogramme. La répartition de l'intensité des images est modélisée par le modèle de mélanges gaussiens (MMG). Les paramètres du MMG sont estimés par l’algorithme «Expectation Maximization». Ensuite, ils sont utilisés pour guider l'évolution des courbes pour atteindre la segmentation des tissus cérébraux. Nous proposons ensuite une amélioration d’un algorithme basé sur les contours actifs orientés région avec la contrainte géométrique. Grâce à la nouvelle expression proposée, il permet de résoudre le problème de stabilité sous-jacente associé à l'algorithme d’origine, et réalise une convergence rapide. Enfin, nous présentons une segmentation de multi-classes en intégrant une segmentation floue dans la méthode level sets. Elle utilise un ensemble d'équations différentielles ordinaires. Chacune d'elles représente une classe à segmenter. Cette approche réduit la complexité de calcul par rapport à l'algorithme multi-phase existant, permettant donc d’accélérer la vitesse de convergence. Toutes les méthodes ont été évaluées avec des images IRM simulées et réelles. Les analyses quantitatives sont données. Les résultats sont très encourageantsThe research goal of this thesis is to develop an automatic segmentation method to segment brain MRI images into different tissues (gray matter, white matter, and cerebrospinal fluid), providing quantitative and precise brain measurements. In this dissertation, we have developed three non-parametric deformable models integrating statistical information and fuzzy information of images to segment the brain into different tissue types from multi types of MRI images. We firstly present a histogram analysis based algorithm, where the intensity distribution of the MRI images is modeled via the mixture Gaussian model (MGM). The parameters of components in MGM are estimated via the Expectation Maximization (EM) algorithm. Then the estimated parameters are used to guide the evolution of the level set curves to achieve the brain tissue segmentation. We then propose an improved algorithm to region-based geometric active contour. Thanks to the new regional term, the new algorithm solves the underlying stability problem associated with the original algorithm, and achieves convergence with less iteration number compared with the original algorithm. Finally, we present a multiclass algorithm by integrating fuzzy segmentation with the level set methods. The algorithm uses a set of ordinary differential equations; each of them represents a class to be segmented. The multiclass algorithm reduces the computational complexity compared with the existing multiphase algorithm, so speeds up the convergence rate. All algorithms are evaluated with simulated and real MRI images, and quantitative analyses are provided. The results are very encouraging
41
Schwaller, Loïc. "Exact Bayesian Inference in Graphical Models : Tree-structured Network Inference and Segmentation." Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLS210/document.
Повний текст джерелаАнотація:
Cette thèse porte sur l'inférence de réseaux. Le cadre statistique naturel à ce genre de problèmes est celui des modèles graphiques, dans lesquels les relations de dépendance et d'indépendance conditionnelles vérifiées par une distribution multivariée sont représentées à l'aide d'un graphe. Il s'agit alors d'apprendre la structure du modèle à partir d'observations portant sur les sommets. Nous considérons le problème d'un point de vue bayésien. Nous avons également décidé de nous concentrer sur un sous-ensemble de graphes permettant d'effectuer l'inférence de manière exacte et efficace, à savoir celui des arbres couvrants. Il est en effet possible d'intégrer une fonction définie sur les arbres couvrants en un temps cubique par rapport au nombre de variables à la condition que cette fonction factorise selon les arêtes, et ce malgré le cardinal super-exponentiel de cet ensemble. En choisissant les distributions a priori sur la structure et les paramètres du modèle de manière appropriée, il est possible de tirer parti de ce résultat pour l'inférence de modèles graphiques arborescents. Nous proposons un cadre formel complet pour cette approche.Nous nous intéressons également au cas où les observations sont organisées en série temporelle. En faisant l'hypothèse que la structure du modèle graphique latent subit un certain nombre de brusques changements, le but est alors de retrouver le nombre et la position de ces points de rupture. Il s'agit donc d'un problème de segmentation. Sous certaines hypothèses de factorisation, l'exploration exhaustive de l'ensemble des segmentations est permise et, combinée aux résultats sur les arbres couvrants, permet d'obtenir, entre autres, la distribution a posteriori des points de ruptures en un temps polynomial à la fois par rapport au nombre de variables et à la longueur de la sérieIn this dissertation we investigate the problem of network inference. The statistical frame- work tailored to this task is that of graphical models, in which the (in)dependence relation- ships satis ed by a multivariate distribution are represented through a graph. We consider the problem from a Bayesian perspective and focus on a subset of graphs making structure inference possible in an exact and e cient manner, namely spanning trees. Indeed, the integration of a function de ned on spanning trees can be performed with cubic complexity with respect to number of variables under some factorisation assumption on the edges, in spite of the super-exponential cardinality of this set. A careful choice of prior distributions on both graphs and distribution parameters allows to use this result for network inference in tree-structured graphical models, for which we provide a complete and formal framework.We also consider the situation in which observations are organised in a multivariate time- series. We assume that the underlying graph describing the dependence structure of the distribution is a ected by an unknown number of abrupt changes throughout time. Our goal is then to retrieve the number and locations of these change-points, therefore dealing with a segmentation problem. Using spanning trees and assuming that segments are inde- pendent from one another, we show that this can be achieved with polynomial complexity with respect to both the number of variables and the length of the series
42
Suzani, Amin. "Automatic vertebrae localization, identification, and segmentation using deep learning and statistical models." Thesis, University of British Columbia, 2014. http://hdl.handle.net/2429/50722.
Повний текст джерелаАнотація:
Automatic localization and identification of vertebrae in medical images of the spine are core requirements for building computer-aided systems for spine diagnosis. Automated algorithms for segmentation of vertebral structures can also benefit these systems for diagnosis of a range of spine pathologies. The fundamental challenges associated with the above-stated tasks arise from the repetitive nature of vertebral structures, restrictions in field of view, presence of spine pathologies or surgical implants, and poor contrast of the target structures in some imaging modalities.
This thesis presents an automatic method for localization, identification, and segmentation of vertebrae in volumetric computed tomography (CT) scans and magnetic resonance (MR) images of the spine. The method makes no assumptions about which section of the vertebral column is visible in the image. An efficient deep learning approach is used to predict the location of each vertebra based on its contextual information in the image. Then, a statistical multi-vertebrae model is initialized by the localized vertebrae from the previous step. An iterative expectation maximization technique is used to register the statistical multi-vertebrae model to the edge points of the image in order to achieve a fast and reliable segmentation of vertebral bodies. State-of-the-art results are obtained for vertebrae localization in a public dataset of 224 arbitrary-field-of-view CT scans of pathological cases. Promising results are also obtained from quantitative evaluation of the automated segmentation method on volumetric MR images of the spine.Applied Science, Faculty of
Electrical and Computer Engineering, Department of
Graduate
43
Möller, Sebastian. "Image Segmentation and Target Tracking using Computer Vision." Thesis, Linköpings universitet, Datorseende, 2011. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-68061.
Повний текст джерелаАнотація:
In this master thesis the possibility of detecting and tracking objects in multispectral infrared video sequences is investigated. The current method with fix-sized rectangles have significant disadvantages. These disadvantages will be solved using image segmentation to estimate the shape of the object. The result of the image segmentation is used to determine the infrared contrast of the object. Our results show how some objects will give very good segmentation, tracking as well as shape detection. The objects that perform best are the flares and countermeasures. But especially helicopters seen from the side, with significant movements, is better detected with our method. The motion of the object is very important since movement is the main component in successful shape detection. This is so because helicopters are much colder than flares and engines. Detecting the presence and position of moving objects is easier and can be done quite successfully even with helicopters. But using structure tensors we can also detect the presence and estimate the position for stationary objects.I detta examensarbete undersöks möjligheterna att detektera och spåra intressanta objekt i multispektrala infraröda videosekvenser. Den nuvarande metoden, som använder sig av rektanglar med fix storlek, har sina nackdelar. Dessa nackdelar kommer att lösas med hjälp av bildsegmentering för att uppskatta formen på önskade mål.Utöver detektering och spårning försöker vi också att hitta formen och konturen för intressanta objekt för att kunna använda den exaktare passformen vid kontrastberäkningar. Denna framsegmenterade kontur ersätter de gamla fixa rektanglarna som använts tidigare för att beräkna intensitetskontrasten för objekt i de infraröda våglängderna. Resultaten som presenteras visar att det för vissa objekt, som motmedel och facklor, är lättare att få fram en bra kontur samt målföljning än vad det är med helikoptrar, som var en annan önskad måltyp. De svårigheter som uppkommer med helikoptrar beror till stor del på att de är mycket svalare vilket gör att delar av helikoptern kan helt döljas i bruset från bildsensorn. För att kompensera för detta används metoder som utgår ifrån att objektet rör sig mycket i videon så att rörelsen kan användas som detekteringsparameter. Detta ger bra resultat för de videosekvenser där målet rör sig mycket i förhållande till sin storlek.
44
Engelbeen, Céline. "The segmentation problem in radiation therapy." Doctoral thesis, Universite Libre de Bruxelles, 2010. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/210107.
Повний текст джерелаАнотація:
The segmentation problem arises in the elaboration of a radiation therapy plan. After the cancer has been diagnosed and the radiation therapy sessions have been prescribed, the physician has to locate the tumor as well as the organs situated in the radiation field, called the organs at risk. The physician also has to determine the different dosage he wants to deliver in each of them and has to define a lower bound on the dosage for the tumor (which represents the minimum amount of radiation that is needed to have a sufficient control of the tumor) and an upper bound for each organ at risk (which represents the maximum amount of radiation that an organ can receive without damaging). Designing a radiation therapy plan that respects these different bounds of dosage is a complex optimization problem that is usually tackled in three steps. The segmentation problem is one of them.Mathematically, the segmentation problem amounts to decomposing a given nonnegative integer matrix A into a nonnegative integer linear combination of some binary matrices. These matrices have to respect the consecutive ones property. In clinical applications several constraints may arise that reduce the set of binary matrices which respect the consecutive ones property that we can use. We study some of them, as the interleaf distance constraint, the interleaf motion constraint, the tongue-and-groove constraint and the minimum separation constraint.
We consider here different versions of the segmentation problem with different objective functions. Hence we deal with the beam-on time problem in order to minimize the total time during which the patient is irradiated. We study this problem under the interleaf distance and the interleaf motion constraints. We consider as well this last problem under the tongue-and-groove constraint in the binary case. We also take into account the cardinality and the lex-min problem. Finally, we present some results for the approximation problem.
/Le problème de segmentation intervient lors de l'élaboration d'un plan de radiothérapie. Après que le médecin ait localisé la tumeur ainsi que les organes se situant à proximité de celle-ci, il doit aussi déterminer les différents dosages qui devront être délivrés. Il détermine alors une borne inférieure sur le dosage que doit recevoir la tumeur afin d'en avoir un contrôle satisfaisant, et des bornes supérieures sur les dosages des différents organes situés dans le champ. Afin de respecter au mieux ces bornes, le plan de radiothérapie doit être préparé de manière minutieuse. Nous nous intéressons à l'une des étapes à réaliser lors de la détermination de ce plan: l'étape de segmentation.
Mathématiquement, cette étape consiste à décomposer une matrice entière et positive donnée en une combinaison positive entière linéaire de certaines matrices binaires. Ces matrices binaires doivent satisfaire la contrainte des uns consécutifs (cette contrainte impose que les uns de ces matrices soient regroupés en un seul bloc sur chaque ligne). Dans les applications cliniques, certaines contraintes supplémentaires peuvent restreindre l'ensemble des matrices binaires ayant les uns consécutifs (matrices 1C) que l'on peut utiliser. Nous en avons étudié certaines d'entre elles comme celle de la contrainte de chariots, la contrainte d'interdiciton de chevauchements, la contrainte tongue-and-groove et la contrainte de séparation minimum.
Le premier problème auquel nous nous intéressons est de trouver une décomposition de la matrice donnée qui minimise la somme des coefficients des matrices binaires. Nous avons développé des algorithmes polynomiaux qui résolvent ce problème sous la contrainte de chariots et/ou la contrainte d'interdiction de chevauchements. De plus, nous avons pu déterminer que, si la matrice donnée est une matrice binaire, on peut trouver en temps polynomial une telle décomposition sous la contrainte tongue-and-groove.
Afin de diminuer le temps de la séance de radiothérapie, il peut être désirable de minimiser le nombre de matrices 1C utilisées dans la décomposition (en ayant pris soin de préalablement minimiser la somme des coefficients ou non). Nous faisons une étude de ce problème dans différents cas particuliers (la matrice donnée n'est constituée que d'une colonne, ou d'une ligne, ou la plus grande entrée de celle-ci est bornée par une constante). Nous présentons de nouvelles bornes inférieures sur le nombre de matrices 1C ainsi que de nouvelles heuristiques.
Finalement, nous terminons par étudier le cas où l'ensemble des matrices 1C ne nous permet pas de décomposer exactement la matrice donnée. Le but est alors de touver une matrice décomposable qui soit aussi proche que possible de la matrice donnée. Après avoir examiné certains cas polynomiaux nous prouvons que le cas général est difficile à approximer avec une erreur additive de O(mn) où m et n représentent les dimensions de la matrice donnée.
Doctorat en Sciences
info:eu-repo/semantics/nonPublished
45
Pal, Chris. "A Probabilistic Approach to Image Feature Extraction, Segmentation and Interpretation." Thesis, University of Waterloo, 2000. http://hdl.handle.net/10012/1049.
Повний текст джерелаАнотація:
This thesis describes a probabilistic approach to imagesegmentation and interpretation. The focus of the investigation is the development of a systematic way of combining color, brightness, texture and geometric features extracted from an image to arrive at a consistent interpretation for each pixel in the image. The contribution of this thesis is thus the presentation of a novel framework for the fusion of extracted image features producing a segmentation of an image into relevant regions. Further, a solution to the sub-pixel mixing problem is presented based on solving a probabilistic linear program. This work is specifically aimed at interpreting and digitizing multi-spectral aerial imagery of the Earth's surface. The features of interest for extraction are those of relevance to environmental management, monitoring and protection. The presented algorithms are suitable for use within a larger interpretive system. Some results are presented and contrasted with other techniques. The integration of these algorithms into a larger system is based firmly on a probabilistic methodology and the use of statistical decision theory to accomplish uncertain inference within the visual formalism of a graphical probability model.
46
Hays, Peter Sipe. "A vector model for analysis, decomposition and segmentation of textures." Thesis, This resource online, 1990. http://scholar.lib.vt.edu/theses/available/etd-06082009-171152/.
Повний текст джерела
47
Parisot, Sarah. "Understanding, Modeling and Detecting Brain Tumors : Graphical Models and Concurrent Segmentation/Registration methods." Phd thesis, Ecole Centrale Paris, 2013. http://tel.archives-ouvertes.fr/tel-00978520.
Повний текст джерелаАнотація:
The main objective of this thesis is the automatic modeling, understanding and segmentation of diffusively infiltrative tumors known as Diffuse Low-Grade Gliomas. Two approaches exploiting anatomical and spatial prior knowledge have been proposed. We first present the construction of a tumor specific probabilistic atlas describing the tumors' preferential locations in the brain. The proposed atlas constitutes an excellent tool for the study of the mechanisms behind the genesis of the tumors and provides strong spatial cues on where they are expected to appear. The latter characteristic is exploited in a Markov Random Field based segmentation method where the atlas guides the segmentation process as well as characterizes the tumor's preferential location. Second, we introduce a concurrent tumor segmentation and registration with missing correspondences method. The anatomical knowledge introduced by the registration process increases the segmentation quality, while progressively acknowledging the presence of the tumor ensures that the registration is not violated by the missing correspondences without the introduction of a bias. The method is designed as a hierarchical grid-based Markov Random Field model where the segmentation and registration parameters are estimated simultaneously on the grid's control point. The last contribution of this thesis is an uncertainty-driven adaptive sampling approach for such grid-based models in order to ensure precision and accuracy while maintaining robustness and computational efficiency. The potentials of both methods have been demonstrated on a large data-set of heterogeneous Diffuse Low-Grade Gliomas. The proposed methods go beyond the scope of the presented clinical context due to their strong modularity and could easily be adapted to other clinical or computer vision problems.
48
Pundlik, Shrinivas J. "Motion segmentation from clustering of sparse point features using spatially constrained mixture models." Connect to this title online, 2009. http://etd.lib.clemson.edu/documents/1252937182/.
Повний текст джерела
49
Thieu, Quang Tung. "Segmentation by convex active contour models : application to skin lesion and medical images." Paris 13, 2013. http://www.theses.fr/2013PA132063.
Повний текст джерела
50
Korkas, Karolos. "Randomised and L1-penalty approaches to segmentation in time series and regression models." Thesis, London School of Economics and Political Science (University of London), 2014. http://etheses.lse.ac.uk/1032/.
Повний текст джерелаАнотація:
It is a common approach in statistics to assume that the parameters of a stochastic model change. The simplest model involves parameters than can be exactly or approximately piecewise constant. In such a model, the aim is the posteriori detection of the number and location in time of the changes in the parameters. This thesis develops segmentation methods for non-stationary time series and regression models using randomised methods or methods that involve L1 penalties which force the coefficients in a regression model to be exactly zero. Randomised techniques are not commonly found in nonparametric statistics, whereas L1 methods draw heavily from the variable selection literature. Considering these two categories together, apart from other contributions, enables a comparison between them by pointing out strengths and weaknesses. This is achieved by organising the thesis into three main parts. First, we propose a new technique for detecting the number and locations of the change-points in the second-order structure of a time series. The core of the segmentation procedure is the Wild Binary Segmentation method (WBS) of Fryzlewicz (2014), a technique which involves a certain randomised mechanism. The advantage of WBS over the standard Binary Segmentation lies in its localisation feature, thanks to which it works in cases where the spacings between change-points are short. Our main change-point detection statistic is the wavelet periodogram which allows a rigorous estimation of the local autocovariance of a piecewise-stationary process. We provide a proof of consistency and examine the performance of the method on simulated and real data sets. Second, we study the fused lasso estimator which, in its simplest form, deals with the estimation of a piecewise constant function contaminated with Gaussian noise (Friedman et al. (2007)). We show a fast way of implementing the solution path algorithm of Tibshirani and Taylor (2011) and we make a connection between their algorithm and the taut-string method of Davies and Kovac (2001). In addition, a theoretical result and a simulation study indicate that the fused lasso estimator is suboptimal in detecting the location of a change-point. Finally, we propose a method to estimate regression models in which the coefficients vary with respect to some covariate such as time. In particular, we present a path algorithm based on Tibshirani and Taylor (2011) and the fused lasso method of Tibshirani et al. (2005). Thanks to the adaptability of the fused lasso penalty, our proposed method goes beyond the estimation of piecewise constant models to models where the underlying coefficient function can be piecewise linear, quadratic or cubic. Our simulation studies show that in most cases the method outperforms smoothing splines, a common approach in estimating this class of models.