No description available.

This paper describes my participation in a MITRE Corporation sponsored computer science clinic project at Harvey Mudd College as my senior project. The goal of the project was to implement a landmark-based pose correction system as a component in a larger, existing face recognition system. The main contribution I made to the project was the implementation of the Active Shape Models (ASM) algorithm; the inner workings of ASM are explained as well as how the pose correction system makes use of it. Included is the most recent draft (as of this writing) of the final report that my teammates and I produced highlighting the year's accomplishments. Even though there are few quantitative results to show because the clinic program is ongoing, our qualitative results are quite promising.

Submitted by JÚLIO HEBER SILVA (julioheber@yahoo.com.br) on 2017-05-11T19:49:56Z No. of bitstreams: 2 Dissertação - Paulo Freire Sobrinho - 2017.pdf: 10088390 bytes, checksum: 93cbaefd8f5c2dc974201729c2e859f7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-05-15T11:30:14Z (GMT) No. of bitstreams: 2 Dissertação - Paulo Freire Sobrinho - 2017.pdf: 10088390 bytes, checksum: 93cbaefd8f5c2dc974201729c2e859f7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-05-15T11:30:14Z (GMT). No. of bitstreams: 2 Dissertação - Paulo Freire Sobrinho - 2017.pdf: 10088390 bytes, checksum: 93cbaefd8f5c2dc974201729c2e859f7 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-04-13 / Contextualization: Technologies like CAD systems, become ideal as a second opinion, that is, to work in partnership with the doctor. For example, PneumoCAD can be used to perform a diagnosis of absence or absence of pneumonia in children 1 to <5 years of age using X-ray radiographs. Problems: However, the use of PneumoCAD requires a region of interest, referring to the lungs, manually. Based on difficulty and difficulty, we sought a solution that was not found in any research, applied to radiographs, involving PneumoCAD, through the active model, having as a patient children between the ages of 1 and <5 years. Proposal: It is then proposed the use of the active model, associated with the technique developed and called nsAlterar in improvement to segmentation based on ns . Materials and Methods: Fifty-six “padrão ouro” radiographs were submitted to MATLAB, in 8 steps, through modified and improved algorithms, as well as implemented support tools, such as: As well as measures of similarity to investigate quantitatively , On an efficiency of all resources employed for the same purpose. Results: With this question, we obtained, after an analysis of the experiments, a taxon of hits for the right of spraying 75.61% and for the left one in 63.41%, in which nsAlterar promoted the improvement in the distributions, even if They were not segmented correctly, through approximations properly. Conclusions: Based on the active model associated with nsAlterar and other resources, it was possible to complement a functionality of the PneumoCAD system, through the use of segmentation in reais, thus contributing to a higher efficiency and better results. / Contextualização: As tecnologias como sistemas CAD, tornam-se ideais como segunda opinião, ou seja, para trabalhar em parceria com o médico. Por exemplo, o PneumoCAD pode ser utilizado para realização do diagnóstico de ausência ou não de pneumonia em crianças de 1 e < 5 anos de idade, através das radiografias de raios-X. Problemática: Entretanto, a utilização do PneumoCAD exige que uma região de interesse, referente aos pulmões, sejam determinadas manualmente. Baseado nesta exigência e dificuldade buscou-se alguma solução que não foi encontrada em nenhuma pesquisa, aplicada a radiografias, envolvendo o PneumoCAD, através do Active Shape Model, tendo como paciente crianças com idade entre 1 e < 5 anos. Proposta: É, então, proposto o uso do Active Shape Model, associado à técnica desenvolvida e denominada nsAlterar em melhora à segmentação baseada no ns . Materiais e Métodos: Foram submetidas, no MATLAB, 56 amostras de radiografias do “padrão-ouro”, em 8 etapas, através de algoritmos modificados e aperfei- çoados, além de ferramentas implementadas de apoio, como: para o treinamento a partir de exemplos; assim como as medidas de similaridades para buscar investigar, de maneira quantitativa, sobre a eficiência de todos os recursos empregados para o mesmo propósito. Resultados: Com isto, foi obtida, após a análise dos experimentos, a taxa de acertos para o pulmão direito em 75,61% e para o esquerdo em 63,41%, em que o nsAlterar promoveu o melhoramento nas distribuições, mesmo as que não foram segmentadas corretamente, através de aproximações de maneira adequada. Conclusões: A partir do Active Shape Model associado ao nsAlterar e demais recursos, foi possível complementar a funcionalidade do sistema PneumoCAD, através do uso da segmentação em situações reais, contribuindo, assim, para a obtenção de maior eficiência e de melhores resultados.

This thesis presents an investigation into Gyrodactylus species recognition, making use of machine learning classification and feature selection techniques, and explores image feature extraction to demonstrate proof of concept for an envisaged rapid, consistent and secure initial identification of pathogens by field workers and non-expert users. The design of the proposed cognitively inspired framework is able to provide confident discrimination recognition from its non-pathogenic congeners, which is sought in order to assist diagnostics during periods of a suspected outbreak. Accurate identification of pathogens is a key to their control in an aquaculture context and the monogenean worm genus Gyrodactylus provides an ideal test-bed for the selected techniques. In the proposed algorithm, the concept of classification using a single model is extended to include more than one model. In classifying multiple species of Gyrodactylus, experiments using 557 specimens of nine different species, two classifiers and three feature sets were performed. To combine these models, an ensemble based majority voting approach has been adopted. Experimental results with a database of Gyrodactylus species show the superior performance of the ensemble system. Comparison with single classification approaches indicates that the proposed framework produces a marked improvement in classification performance. The second contribution of this thesis is the exploration of image processing techniques. Active Shape Model (ASM) and Complex Network methods are applied to images of the attachment hooks of several species of Gyrodactylus to classify each species according to their true species type. ASM is used to provide landmark points to segment the contour of the image, while the Complex Network model is used to extract the information from the contour of an image. The current system aims to confidently classify species, which is notifiable pathogen of Atlantic salmon, to their true class with high degree of accuracy. Finally, some concluding remarks are made along with proposal for future work.

Osteoporosis, associated with reduced bone mineral density and structural degeneration, greatly increases the risk of fragility fracture. Magnetic resonance imaging (MRI) has been applied to central skeletal sites including the proximal femur due to its non-ionizing radiation. A major challenge of volumetric bone imaging of the hip is the selection of regions of interest (ROIs) for computation of regional bone measurements. To address this issue, an MRI-based active shape model (ASM) of the human proximal femur is applied to automatically generate ROIs. The challenge in developing the ASM for a complex three-dimensional (3-D) shape lies in determining a large number of anatomically consistent landmarks for a set of training shapes. This thesis proposes a new method of generating the proximal femur ASM, where two types of landmarks, namely fiducial and secondary landmarks, are used. The method consists of—(1) segmentation of the proximal femur bone volume, (2) smoothing the bone surface, (3) drawing fiducial landmark lines on training shapes, (4) drawing secondary landmarks on a reference shape, (5) landmark mesh generation on the reference shape using both fiducial and secondary landmarks, (6) generation of secondary landmarks on other training shapes using the correspondence of fiducial landmarks and an elastic deformation of the landmark mesh, (7) computation of the active shape model. A proximal femur ASM has been developed using hip MR scans of 45 post-menopausal women. The results of secondary landmark generation were visually satisfactory, and no topology violation or notable geometric distortion artifacts were observed. Performance of the method was examined in terms of shape representation errors in a leave-one-out test. The mean and standard deviation of leave-one-out shape representation errors were 0.34mm and 0.09mm respectively. The experimental results suggest that the framework of fiducial and secondary landmarks allows reliable computation of statistical shape models for complex 3-D anatomic structures.

Osteoporosis is a bone disease associated with fracture risk. Accurate assessments of fracture risk, guidelines to initiate preventive intervention, and monitoring treatment response are of paramount importance in public health. Clinically, osteoporosis is defined by low bone mineral density, which explains 65-75% of the variance in bone stiffness. The remaining variability is due to the cumulative and synergistic effects of various factors, including trabecular bone micro-architecture. Osteoporostic imaging is critically important in identifying fracture risks for planning of therapeutic intervention and monitoring response to treatments. In this work, quantitative analysis of trabecular bone micro-architecture using volumetric imaging techniques and computational biomechanical simulation through finite element modeling (FEM) are applied on in vivo imaging for various human studies. The ability of imaging methods in characterizing trabecular bone micro-architecture was experimentally examined using MRI and multi-row detector CT. They were found suitable for cross-sectional and longitudinal studies in monitoring changes of trabecular micro-architectural quality in clinical research. A framework which consists of robust segmentation of in vivo images and quality mesh generator, was constructed for FEM analysis. The framework was experimentally demonstrated effcient and effective to predict bone strength under limited spatial resolution. The ability of distinguishing bone strengths of different groups were evaluated on various human studies. And the relation between FEM and image-based micro-architectural measures was explored. Quantitative analysis supports the hypothesis that trabecular bone have distinct structural properties in different anatomic sites and the osteoporosis related change of the micro-architecture also varies. It highlight the importance of standardizing the definition of bone scan locations and the segmentation of such well-defined regions. A shape modeling method was proposed to solve the problem and its application in human proximal femur using MRI were presented. The method was compared with manual segmentation and found highly accurate. Together with tools developed for quantitative analysis, this work facilitates future researches of trabecular bone micro-architecture in different anatomic sites.

[pt] As expressões faciais transmitem muita informação sobre um indivíduo, tornando a capacidade de interpretá-las uma tarefa muito importante, com aplicações em diversas áreas, tais como Interação Homem Máquina, Jogos Digitais, storytelling interativo e TV/Cinema digital. Esta dissertação discute o processo de reconhecimento de emoções em tempo real usando ASM (Active Shape Model) e SVM (Support Vector Machine) e apresenta uma comparação entre duas formas comumente utilizadas na etapa de extração de atributos: faces neutra e média. Como não existe tal comparação na literatura, os resultados apresentados são valiosos para o desenvolvimento de aplicações envolvendo expressões de emoção em tempo real. O presente trabalho considera seis tipos de emoções: felicidade, tristeza, raiva, medo, surpresa e desgosto. / [en] The facial expressions provide a high amount of information about a person, making the ability to interpret them a high valued task that can be used in several fields of Informatics such as Human Machine Interface, Digital Games, interactive storytelling and digital TV/Cinema. This dissertation discusses the process of recognizing emotions in real time using ASM (Active Shape Model) and SVM (Support Vector Machine) and presents a comparison between two commonly used ways when extracting the attributes: neutral face and average. As such comparison can not be found in the literature, the results presented are valuable to the development of applications that deal with emotion expression in real time. The current study considers six types of emotions: happiness, sadness, anger, fear, surprise and disgust.

Optimal surface finding (OSF), a graph-based optimization approach to image segmentation, represents a powerful framework for medical image segmentation and analysis. In many applications, a pre-segmentation is required to enable OSF graph construction. Also, the cost function design is critical for the success of OSF. In this thesis, two issues in the context of OSF segmentation are addressed. First, a robust model-based segmentation method suitable for OSF initialization is introduced. Second, an OSF-based segmentation refinement approach is presented. For segmenting complex anatomical structures (e.g., lungs), a rough initial segmentation is required to apply an OSF-based approach. For this purpose, a novel robust active shape model (RASM) is presented. The RASM matching in combination with OSF is investigated in the context of segmenting lungs with large lung cancer masses in 3D CT scans. The robustness and effectiveness of this approach is demonstrated on 30 lung scans containing 20 normal lungs and 40 diseased lungs where conventional segmentation methods frequently fail to deliver usable results. The developed RASM approach is generally applicable and suitable for large organs/structures. While providing high levels of performance in most cases, OSF-based approaches may fail in a local region in the presence of pathology or other local challenges. A new (generic) interactive refinement approach for correcting local segmentation errors based on the OSF segmentation framework is proposed. Following the automated segmentation, the user can inspect the result and correct local or regional segmentation inaccuracies by (iteratively) providing clues regarding the location of the correct surface. This expert information is utilized to modify the previously calculated cost function, locally re-optimizing the underlying modified graph without a need to start the new optimization from scratch. For refinement, a hybrid desktop/virtual reality user interface based on stereoscopic visualization technology and advanced interaction techniques is utilized for efficient interaction with the segmentations (surfaces). The proposed generic interactive refinement method is adapted to three applications. First, two refinement tools for 3D lung segmentation are proposed, and the performance is assessed on 30 test cases from 18 CT lung scans. Second, in a feasibility study, the approach is expanded to 4D OSF-based lung segmentation refinement and an assessment of performance is provided. Finally, a dual-surface OSF-based intravascular ultrasound (IVUS) image segmentation framework is introduced, application specific segmentation refinement methods are developed, and an evaluation on 41 test cases is presented. As demonstrated by experiments, OSF-based segmentation refinement is a promising approach to address challenges in medical image segmentation.

Dans cette thèse, nous nous intéressons au problème de la segmentation de portraits numériques. Nous appelons portrait numérique la photographie d’une personne avec un cadre allant grossièrement du gros plan au plan poitrine. Le problème abordé dans ce travail est un cas spécifique de la segmentation d’images où il s’agit notamment de définir précisément la frontière de la région « cheveux ». Ce problème est par essence très délicat car les attributs de la région « cheveux » (géométrie, couleur, texture) présentent une grande variabilité à la fois entre les personnes et au sein de la région. Notre cadre applicatif est un système d’« essayage virtuel » de lunettes à destination du grand public, il n’est pas possible de contrôler les conditions de prise de vue comme l’éclairage de la scène ou la résolution des images, ce qui accroît encore la diculté du problème. L’approche proposée pour la segmentation de portraits numériques est une approche du plus grossier au plus fin procédant par étapes successives. Nous formulons le problème comme celui d’une segmentation multi-régions, en introduisant comme « régions secondaires », les régions adjacentes à la région « cheveux » , c.-à-d. les régions « peau » et « fond ». La méthode est fondée sur l’apparence (appearance-based method) et a comme spécificité le fait de déterminer les descripteurs de régions les plus adaptés à partir d’une base d’images d’apprentissage et d’outils statistiques. À la première étape de la méthode, nous utilisons l’information contextuelle d’un portrait numérique — connaissances a priori sur les relations spatiales entre régions— pour obtenir des échantillons des régions « cheveux », « peau » et « fond ». L’intérêt d’une approche fondée sur l’apparence est de pouvoir s’adapter à la fois aux conditions de prises de vue ainsi qu’aux attributs de chaque régions. Au cours de cette étape, nous privilégions les modèles de forme polygonaux couplés aux contours actifs pour assurer la robustesse du modèle. Lors de la seconde étape, à partir des échantillons détectés à l’étape précédente, nous introduisons un descripteur prenant en compte l’information de couleur et de texture. Nous proposons une segmentation grossière par classification en nous appuyant à nouveau sur l’information contextuelle : locale d’une part grâce aux champs de Markov, globale d’autre part grâce à un modèle a priori de segmentation obtenu par apprentissage qui permet de rendre les résultats plus robustes. La troisième étape ane les résultats en définissant la frontière des « cheveux » comme une région de transition. Cette dernière contient les pixels dont l’apparence provient du mélange de contributions de deux régions (« cheveux »et « peau » ou «fond »). Ces deux régions de transition sont post-traitées par un algorithme de «démélange » (digital matting) pour estimer les coecients de transparence entre « cheveux » et « peau », et entre « cheveux » et « fond ». À l’issue de ces trois étapes, nous obtenons une segmentation précise d’un portrait numérique en trois « calques », contenant en chaque pixel l’information de transparence entre les régions « cheveux », « peau » et « fond ». Les résultats obtenus sur une base d’images de portraits numériques ont mis en évidence les bonnes performances de notre méthode. / In this thesis, we are interested in the problem of the segmentation of digital portraits. We call digital portrait the photography of a person with a frame roughly ranging from the close-up to the chest plane. The problem addressed in this work is a specific case of the segmentation of images where it is especially necessary to define precisely the border of the "hair" region. This problem is inherently very delicate because the attributes of the "hair" region (geometry, color, texture) present an important variability between people and within the region. Our application is a system of "virtual fitting" of glasses for the general audience, it is not possible to control the shooting conditions such as stage lighting or image resolution, which increases the difficulty of the problem. The approach proposed for the segmentation of digital portraits is an approach « coarse to fine », proceeding in successive stages. We formulate the problem as a multi-region segmentation, introducing as "secondary regions" regions adjacent to the "hair" region, ie, the "skin" and "background" regions. The method is based on appearance-based method and has as a specificity the determination of the descriptors of regions most adapted from a database of learning and statistical tools. In the first step of the method, we use the contextual information of a Digital portrait - a priori knowledge about the spatial relations between regions - to obtain samples of the regions "hair", "skin" and "background". The value of an appearance-based approach is to be able to adapt to both the shooting conditions and the attributes of each region. During this stage, we prefer polygonal shape models coupled with active contours to ensure the robustness of the model. In the second step, from the samples detected in the previous step, we introduce a descriptor taking into account the color and texture information. We propose a rough segmentation by classification by relying on the contextual information: local on the one hand thanks to the Markov fields, global on the other hand thanks to an a priori model of segmentation obtained by learning which il allow to obtain robust results. The third stage refines the results by defining the border of "hair" as a transition region. This région contains the pixels whose appearance comes from the mixture of contributions of two regions ("hair" and "skin" or "background"). These two transition regions are post-processed by a digital matting algorithm to estimate the coefficients of transparency between "hair" and "skin", and between "hair" and "background". At the end of these three steps, we obtain a precise segmentation of a digital portrait into three "layers", containing in each pixel the information of transparency between the regions "hair", "skin" and "background". The results obtained on the basis of images of digital portraits have highlighted the good performance of our method.

As imagens são sinais que possuem muita informação. Os objetos representados em ima- gens podem sofrer deformações fazendo com que suas características mudem, o que difi- culta o reconhecimento do objeto. A chave para o sucesso de um sistema de reconheci- mento de padrões em imagens é escolher adequadamente a sua abordagem e um modelo para as feições presentes nas imagens. Uma das dificuldades é extrair e selecionar as feições que são mais discriminantes entre as diferentes classes usadas para modelar um objeto. Os modelos ativos de formas (Active Shape Models ASM) adaptam-se às defor- mações de um objeto. O objeto a ser modelado, pode ser representado com um modelo de pontos distribuídos (Point Distribution Models PDM). O PDM consiste de pontos de interesse ou marcadores, que permitem a extração de feições em localizações específicas do objeto. Após tratar a incerteza da localização e oclusão dos marcadores é possível ex- trair as feições mais representativas, obtendo-se um desempenho alto em termos da taxa de reconhecimento. Nesta tese são introduzidas novas formas para extrair e selecionar feições com modelos ativos de formas, que melhoram a taxa de classificação onde há objetos deformáveis. Esta tese é inovadora no sentido de aperfeiçoar o uso de ASMs na classificação de faces humanas, e na sua aplicação no monitoramento visual de outros tipos de objetos deformáveis. / Images are signals that have a lot of information. The objects depicted in images may suf- fer deformations causing changes in their characteristics, which hinders the recognition of the object. The key to the success of a system of pattern recognition in images is to choose properly your approach and a model for the features in the images. One difficulty is to extract and select the features that are most discriminating between different classes used to model an object. The Active Shape Models (ASM) adapt to deformations of an object. The object to be modeled can be represented with a Points Distribution Model (PDM). The PDM consists of points of interest or landmarks that allow the extraction of features in specific locations of the object. After treating the uncertainty of the location and occlu- sion of the landmarks it is possible to extract the most representative features, obtaining a high performance in terms of recognition rate. This thesis introduces new ways to extract and select features with ASMs, which improve the classification rate where deformable objects are present. This thesis is innovative in the sense that improves the use of ASMs in the classification of human faces, and that can be applied in visual monitoring of other types of deformable objects.