Glaucoma is a common disease that is difficult to diagnosis early using only visual field tests. Current research indicates that determination of retinal nerve fiber layer thickness (RNFLT) can serve as an early indicator of glaucoma [1]. RNFLT is measured by segmenting non-invasive optical coherence tomography images. However, high speckle noise and presence of artifacts in the images cause traditional layer detection and segmentation methods to fail. Multiphase active contours segmentation methods utilize region intensity and shape terms to produce multiple continuous boundaries simultaneously in noisy environments. A 2D and 3D multiphase active contours based algorithm was created to segment synthetic and real human retina OCT images. The 2D multiphase algorithm segmented eight simultaneous layers with a 3.14% mean A-scan error rate per layer. The 3D approach performed qualitatively accurate segmentation of a 20 image stack simultaneously. In an artificial, high-noise image stack the incorporation of more pixels per layer allowed improved segmentation using the 3D algorithm over the 2D. These results indicate that 2D and 3D multiphase active contours algorithms can be used to accurately segment retina layers. With further development to reduce computation time and automate initialization, these algorithms could be used to provide close to real-time clinical retinal image segmentation. / text

This thesis develops image segmentation methods for the application of automated cervical cancer screening. The traditional approach to automating this task has been to emulate the human method of screening, where every one of the hundreds of thousands of cells on each slide is analysed for abnormality. However, due to the complexity of cervical smear images and the low error tolerance imposed upon the segmentation stage, only limited success has previously been found. A different approach is to detect malignancy associated changes (MACs) in a relatively small sample of the total population of cells. Under this paradigm, the requirement to segment every cell is loosened, but delineation accuracy and error checking become essential. Following a review of generic and cervical smear segmentation, it is concluded that prior work on the traditional approach to automation is not suitable for a MACs solution. However, the previously proposed framework of a dual-magnification system is found to be relevant and is therefore adopted. Here, scene images are first captured at low resolution in order to rapidly locate the cells on a slide. Cells that are deemed to be suitable for further analysis are then imaged at high resolution for the more accurate segmentation of their nuclei. A water immersion algorithm is developed for low resolution scene segmentation. This method achieves a rapid and robust initial segmentation of the scene without the requirement of incorporating extensive a priori knowledge of the image objects. A global minimum searching contour is presented as a top-down method for segmenting the high resolution cell nucleus images where the image objects are well characterised by shape and appearance. This latter method is tested upon 20,000 images and found to achieve an accurate segmentation rate of 99.47%. An error checking method, that uses segmentation stability as an indicator of segmentation success, is developed that is capable of detecting 100% of the failures of the nucleus segmenter, at the expense of discarding only 9% of the data. Throughout this work, contemporary issues in the field of generic image segmentation are presented and some of these are addressed for the cervical smear application. Finally, an avenue of future work is proposed which may lead to the much wider proliferation of computer vision solutions to everyday problems.

In this dissertation, novel variational optical-flow and active-contour methods are investigated to address challenging problems in computer vision and shape-based imaging. Starting from traditional applications of these methods in computer vision, such as object segmentation, tracking, and detection, this research subsequently applies similar active contour techniques to the realm of shape-based imaging, which is an image reconstruction technique estimating object shapes directly from physical wave measurements. In particular, the first and second part of this thesis deal with the following two physically inspired computer vision applications. Optical Flow for Vision-Based Flame Detection: Fire motion is estimated using optimal mass transport optical flow, whose motion model is inspired by the physical law of mass conservation, a governing equation for fire dynamics. The estimated motion fields are used to first detect candidate regions characterized by high motion activity, which are then tracked over time using active contours. To classify candidate regions, a neural net is trained on a set of novel motion features, which are extracted from optical flow fields of candidate regions. Coupled Photo-Geometric Object Features: Active contour models for segmentation in thermal videos are presented, which generalize the well-known Mumford-Shah functional. The diffusive nature of heat processes in thermal imagery motivates the use of Mumford-Shah-type smooth approximations for the image radiance. Mumford-Shah's isotropic smoothness constraint is generalized to anisotropic diffusion in this dissertation, where the image gradient is decomposed into components parallel and perpendicular to level set curves describing the object's boundary contour. In a limiting case, this anisotropic Mumford-Shah segmentation energy yields a one-dimensional ``photo-geometric'' representation of an object which is invariant to translation, rotation and scale. These properties allow the photo-geometric object representation to be efficiently used as a radiance feature; a recognition-segmentation active contour energy, whose shape and radiance follow a training model obtained by principal component analysis of a training set's shape and radiance features, is finally applied to tracking problems in thermal imagery. The third part of this thesis investigates a physics-driven active contour approach for shape-based imaging. Adjoint Active Contours for Shape-Based Imaging: The goal of this research is to estimate both location and shape of buried objects from surface measurements of waves scattered from the object. These objects' shapes are described by active contours: A misfit energy quantifying the discrepancy between measured and simulated wave amplitudes is minimized with respect to object shape using the adjoint state method. The minimizing active contour evolution requires numerical forward scattering solutions, which are obtained by way of the method of fundamental solutions, a meshfree collocation method. In combination with active contours being implemented as level sets, one obtains a completely meshfree algorithm; a considerable advantage over previous work in this field. With future applications in medical and geophysical imaging in mind, the method is formulated for acoustic and elastodynamic wave processes in the frequency domain.

Real-time face recognition has recently become available for the government and industry due to developments in face recognition algorithms, human head detection algorithms, and faster/low cost computers. Despite these advances, however, there are still some critical issues that affect the performance of real-time face recognition software. This paper addresses the problem of off-centered and out-of-pose faces in pictures, particularly in regard to the eigenface method for face recognition. We first demonstrate how the representation of faces by the eigenface method, and ultimately the performance of the software depend on the location of the eyes in the pictures. The eigenface method for face recognition is described: specifically, the creation of a face basis using the singular value decomposition, the reduction of dimension, and the unique representation of faces in the basis. Two different approaches for aligning the eyes in images are presented. The first considers the rotation of images using the orthogonal Procrustes Problem. The second approach looks at locating features in images using energy-minimizing active contours. We then conclude with a simple and fast algorithm for locating faces in images. Future research is also discussed. / Master of Science

Light Detection And Ranging (LiDAR), as well as many other applications and sensors, involve segmenting sparse sets of points (point clouds) for which point density is the only discriminating feature. The segmentation of these point clouds is challenging for several reasons, including the fact that the points are not associated with a regular grid. Moreover, the presence of noise, particularly impulsive noise with varying density, can make it difficult to obtain a good segmentation using traditional techniques, including the algorithms that had been developed to process LiDAR data. This dissertation introduces novel algorithms and frameworks based on statistical techniques and image analysis in order to segment and extract surfaces from sparse noisy point clouds. We introduce an adaptive method for mapping point clouds onto an image grid followed by a contour detection approach that is based on an enhanced version of region-based Active Contours Without Edges (ACWE). We also proposed a noise reduction method using Bayesian approach and incorporated it, along with other noise reduction approaches, into a joint framework that produces robust results. We combined the aforementioned techniques with a statistical surface refinement method to introduce a novel framework to detect ground and canopy surfaces in micropulse photon-counting LiDAR data. The algorithm is fully automatic and uses no prior elevation or geographic information to extract surfaces. Moreover, we propose a novel segmentation framework for noisy point clouds in the plane based on a Markov random field (MRF) optimization that we call Point Cloud Densitybased Segmentation (PCDS). We also developed a large synthetic dataset of in plane point clouds that includes either a set of randomly placed, sized and oriented primitive objects (circle, rectangle and triangle) or an arbitrary shape that forms a simple approximation for the LiDAR point clouds. The experiment performed on a large number of real LiDAR and synthetic point clouds showed that our proposed frameworks and algorithms outperforms the state-of-the-art algorithms in terms of segmentation accuracy and surface RMSE. / Ph. D. / The increasing concerns about the global warming have raised the interest about studying and understanding the global ecosystem components including the carbon cycle. The interaction between forests and earth atmosphere is one major component of the global carbon cycle. Thus, quantifying the global forest biomass is an important factor in studying carbon cycle and its dynamics. Therefore repeated large-scale estimates of forest biomass are critically important. LIDAR (Light Detection and Ranging) is a active remote sensing method that uses light in the form of a pulsed laser to measure ranges and distances based on the time-of-flight concept (similar to radar systems). LiDAR systems can generate precise, three-dimensional information about the shape of the Earth and its surface characteristics. Therefore, LiDAR remote sensing is much more suitable for forest studies than photogrammetry because of the laser’s ability to penetrate tree crowns allowing the system to find ground returns under dense canopies. This property allows us to estimate tree heights which is a major factor for estimating the forest biomass. In order to track forest biomass changes at the global scale, recurring high-altitude observations are needed. Satellite-based LiDAR systems can provide these observations, although no such systems are currently operational. The situation will change with the launch of NASAs ICESat-2, which is planned for July 2017. However, although LiDAR technology allows for rapid and inexpensive measurements over broad geographical areas, ICESat-2 will be equipped with a new sensor known as photon-counting micropulse LiDAR system. This new LiDAR technology is expected to produce measurements that include high levels of noise. The data produced by this sensor will be in the form of a cloud of points in which the signal points are expected to be much more dense than noise points. Analysis of data from the ICESat-2 satellite will therefore need to be robust with respect to noise, as well as fast and automatic because of the large quantity of data that will be generated. The problem of segmentation in point clouds is challenging for several reasons, including the fact that the points are not associated with a regular grid, as is the case with most image data. Moreover, the presence of noise particularly impulsive noise with varying density, can make it difficult to obtain a good segmentation using traditional techniques, including the algorithms that had been developed to process LiDAR data. This dissertation introduces novel algorithms and approaches based on statistical techniques and image analysis in order to segment sparse noisy point clouds to extract contours and surfaces in order to detect meaningful measurements and information.

Técnicas de modelagem baseada em imagens têm recebido considerável atenção da comunidade de visualização computacional devido ao potencial de criar cenas realistas a partir de um pequeno conjunto de imagens bi-dimensionais. Entretanto, a qualidade dos modelos gerados pelas ferramentas atualmente disponíveis é extremamente dependente de entradas fornecidas pelo usuário. Este trabalho propõe a execução do projeto de uma ferramenta de auxílio para sistemas de modelagem baseada em imagens que utiliza o conceito de contornos ativos para aumentar o grau de automação do processo de localização do contorno do objeto presente na fotografia, que servirá de guia para a posterior localização dos vértices desse objeto. Através desta abordagem, figuras geométricas mais simples, como pirâmides e hexaedros, puderam ser reconstruídas após a recuperação das coordenadas de seus vértices / Image Based Modelling techniques has received considerable attention from the computer vision community due to the potential to create realistic scenes from some bi-dimensional images. However, the model?s quality generated by the tools available nowadays is extremely dependent on entries provided by the user. This work proposes the execution of a help tool project for image based modelling systems that uses the active contours concept to increase the process automation degree of locating the contour of an object in the image, which will guide the vertex location process of this object. Through this approach, simple geometric figures, as pyramids and squares, could be reconstructed after the vertex coordinates recuperation

Este trabalho representa continuidade no desenvolvimento de trabalhos na área de neurociência computacional, em particular na área de neuromorfometria e no relacionamento da forma-função. Os objetivos principais são a investigação e a simulação de modelos dinâmicos para o desenvolvimento de células neurais, e a caracterização da sua morfometria em termos de atributos. A tese apresenta um histórico sobre a neurociência, e uma breve revisão sobre a biologia do neurônio e sobre fatores que influenciam na variação na sua forma. Seguimos com a apresentação dos principais modelos computacionais de simulação neural, funcionais e de crescimento neural, com uma descrição mais detalhada de um modelo de crescimento baseado na atuação do cálcio como agente morfogênico e também na polimerização de actinas. Como uma introdução à modelagem neural, discutimos técnicas computacionais de evolução de contornos que podem ser utilizadas na simulação do desenvolvimento neural, propagação de frentes e contornos ativos. Apresentamos também medidas neuromorfométricas tais como a dimensão fractal multiescala, e medidas extraídas a partir do esqueleto da imagem do neurônio, tais como largura, espessura, número de ramos e curvatura das ramificações. Apresentamos os resultados obtidos em diferentes hipóteses de desenvolvimento de células neurais. Foram propostos crescimentos baseados na normal (velocidade na direção normal a curva), convolução, thin plate splines e dinâmica da polimerização da actina. Além disso, foi proposta uma nova abordagem para a evolução da membrana neural baseada em contornos, utilizando a formulação de contornos ativos sob a ação do campo elétrico externo e a curvatura da forma, o que possibilitou a geração de estruturas com características muito semelhantes a do neurônio, inclusive com ramificações. Finalizamos o trabalho apresentando os resultados e conclusões obtidas para os modelos de desenvolvimento. / In this thesis we report the investigation and simulation of dynamic models of neural growing, and their characterization using shape features, considering the form function relationship and neuromorphometry. The thesis begins by presenting an overview about neuroscience, neural cell biology and the biological factors that affects the neuron form developments, followed by the presentation of computational neuronal models based on electrophisiological measures and development models of internal structures as actin and microtubules. Special attention is devoted to a neuron growth model based on calcium as a morphogen, whose main characteristic is its electric activity at the membrane. Regarding mathematical models of neural development, two different approaches of contour evolutions are presented, Level Set Methods and Active Contours. Some neuromorphometric measures are implemented and discussed as features for classification and neural evolution, including the multiscale fractal dimension, and dendrite measurements are obtained by using neuron skeletons. In agreement with biological form influences, some hypotheses about development of neuron growth are proposed based on evolution rules, such as: normal evolution (based in normal velocity), convolution, thin plate splines and actin polimerization. A new approach about neuron development is also proposed: a contour based technique that makes use of active contour formulation, Snake Balloon, where the membrane velocity and direction suffers influences of internal and external factors, such as electrical field with diferent geometries, and contour curvature. Both hypotheses are in accordance with the biological factors that influences the neuron form. The simulation produces similar neuron-like structures, even with ramification of certain dendrites

Cette thèse concerne la segmentation d’images HRpQCT et l’évaluation d’indices morphologiques de l’os cortical pour le diagnostic de l’ostéoporose et la prédiction des fractures osseuses. Dans un premier temps,deux méthodes sont proposées pour la segmentation de l’os cortical. La première utilise une nouvelle approche des contours actifs basée sur la logique floue suivie d’une nouvelle technique de remplissage développée pour imiter le comportement des opérateurs pour séparer l’os cortical de l’os trabéculaire. La deuxième approche est une technique 3D à double contours actifs combinant à la fois les informations locales le long et entre les deux contours. Les deux approches de segmentation sont comparées à celles de l’état de l’art afin de valider leurs performances. Dans un second temps, différents indices extraits de l’os cortical sont utilisés pour déterminer leur potentiel de prédiction des fractures ostéoporotiques. Les résultats obtenus montent que l’analyse globale de l’os cortical masque des variations potentiellement importantes.Par conséquent, une décomposition régionale de l’enveloppe corticale est proposée afin d’améliorer la prédiction du risque fracturaire. / This thesis concerns the segmentation of HRpQCT images and the evaluation of the cortical bone parameters for the osteoporosis characterization and the fracture prediction. Firstly, two approaches were proposed to segment the cortical bone. The first uses a new fuzzy energy active contours approach followed by a new filling technique designed to mimic the behaviour of clinicians while extracting the cortical bone from the trabecularone. The second approach is a local based 3D dual active contours approach proposed to separate between three regions constituting the image. To move, this approach combines the local information along each point in the two contours conjointly with the information between them. The segmentation results of these approaches were confronted to the state of the art methods to validate their performance. Secondly,different parameters were extracted from the segmented cortical bone to monitor the association of these parameters with the osteoporotic fracture prediction. Global analysis of the cortical bone obscures potentially important regional variations. Therefore, regional cortical decomposition was proposed to illustrate that cortical sub-regions could improve the evaluation of fracture risk than the global analysis of the cortical bone.

Técnicas de modelagem baseada em imagens têm recebido considerável atenção da comunidade de visualização computacional devido ao potencial de criar cenas realistas a partir de um pequeno conjunto de imagens bi-dimensionais. Entretanto, a qualidade dos modelos gerados pelas ferramentas atualmente disponíveis é extremamente dependente de entradas fornecidas pelo usuário. Este trabalho propõe a execução do projeto de uma ferramenta de auxílio para sistemas de modelagem baseada em imagens que utiliza o conceito de contornos ativos para aumentar o grau de automação do processo de localização do contorno do objeto presente na fotografia, que servirá de guia para a posterior localização dos vértices desse objeto. Através desta abordagem, figuras geométricas mais simples, como pirâmides e hexaedros, puderam ser reconstruídas após a recuperação das coordenadas de seus vértices / Image Based Modelling techniques has received considerable attention from the computer vision community due to the potential to create realistic scenes from some bi-dimensional images. However, the model?s quality generated by the tools available nowadays is extremely dependent on entries provided by the user. This work proposes the execution of a help tool project for image based modelling systems that uses the active contours concept to increase the process automation degree of locating the contour of an object in the image, which will guide the vertex location process of this object. Through this approach, simple geometric figures, as pyramids and squares, could be reconstructed after the vertex coordinates recuperation

As doenÃas cardÃacas constituem a principal causa de mortalidade em paÃses desenvolvidos e ocupam uma posiÃÃo de destaque em paÃses em desenvolvimento, sendo no Brasil a segunda causa de morte. Na busca por sua identificaÃÃo, diversos exames podem ser feitos, dentre eles o ECG (eletrocardiograma), medicina nuclear e o ecocardiograma de esforÃo (ECE). Este Ãltimo à preferÃvel por ser um mÃtodo de baixo custo, comparando-se com medicina nuclear, alÃm de ser um mÃtodo nÃo evasivo. Por estas razÃes à muito utilizado no diagnÃstico preciso de isquemia (perda de elasticidade muscular), inclusive na sua intensidade. Entretanto, o diagnÃstico por ECE à realizado por uma anÃlise visual de um vÃdeo por um especialista, portanto subjetivo. Esta dissertaÃÃo descreve um mÃtodo de segmentaÃÃo do ventrÃculo esquerdo em ecocardiogramas, utilizando-se de contornos ativos (snakes) na tentativa de tornar o ECE o mais objetivo possÃvel, permitindo uma medida automÃtica do volume do ventrÃculo esquerdo atravÃs da anÃlise do ECE a fim de detectar a isquemia. TÃcnicas de eliminaÃÃo de ruÃdo speckle sÃo implementadas e confrontadas, pois o ECE à sempre contaminado por este tipo de ruÃdo, resultando em uma anÃlise visual de difÃcil percepÃÃo. Tais tÃcnicas utilizam a transformada wavelets na construÃÃo dos filtros, sendo entÃo, implementados e avaliados quatro diferentes algoritmos, tomando-se como parÃmetro o tempo de processamento e a relaÃÃo sinal/ruÃdo. Desenvolvesse tambÃm uma tÃcnica de detecÃÃo do ventrÃculo esquerdo, usando o mÃtodo dos contornos ativos (snakes). Resultados parciais sÃo obtidos, permitindo uma detecÃÃo de bordas do miocÃrdio interno representando as paredes do ventrÃculo esquerdo. / The cardiac illnesses represent the most of death cause in developed countries and one of the commonest in developing countries, which is the case of Brazil. In order to detect them, different existing methods can be applied. Among them, the ECG (electrocardiogram), methods developed from nuclear medicine and echocardiogram of effort (ECE). This last one is preferred due to its low cost, in comparison with the nuclear medicine and its non-invasive ability. For these reasons, ECE is used to diagnostic the isquemia (muscular elasticity loss). But the diagnostic is done by a specialist through the visualization of a video, and is consequently subjective. This master degree dissertation describes a segmentation method of the left ventricle in ecocardiograms, using the active contours model (snake), in order to the ECE becomes more objective, allowing an automatic measure of left ventricle and consequently the detection of isquemia. Denoising techniques are implemented because this kind of image is corrupted by the speckle noise, what is harmful for analises. Wavelet-based technics are developed and four algorithms are compared in measuring the time of execution and the resulting signal to noise ratio. From the resulting denoised images, a technique for left ventricle border detection is developed using the snake methods. Partial results show an effective result for the detection of intern myocardium borders which represent the left ventricle.