A segmentação de imagens é um problema muito importante em visão computacional, no qual uma imagem é dividida em regiões relevantes, tal como para isolar objetos de interesse de uma dada aplicação. Métodos de segmentação baseados na transformada imagem-floresta (IFT, Image Foresting Transform), com funções de conexidade monotonicamente incrementais (MI) têm alcançado um grande sucesso em vários contextos. Na segmentação interativa de imagens, na qual o usuário pode especificar o objeto desejado, novas sementes podem ser adicionadas e/ou removidas para corrigir a rotulação até conseguir a segmentação esperada. Este processo gera uma sequência de IFTs que podem ser calculadas de modo mais eficiente pela DIFT (Differential Image Foresting Transform). Recentemente, funções de conexidade não monotonicamente incrementais (NMI) têm sido usadas com sucesso no arcabouço da IFT no contexto de segmentação de imagens, permitindo incorporar informações de alto nível, tais como, restrições de forma, polaridade de borda e restrição de conexidade, a fim de customizar a segmentação para um dado objeto desejado. Funções não monotonicamente incrementais foram também exploradas com sucesso na geração de superpixels, via sequências de execuções da IFT. Neste trabalho, apresentamos um estudo sobre a Transformada Imagem-Floresta Diferencial no caso de funções NMI. Nossos estudos indicam que o algoritmo da DIFT original apresenta uma série de inconsistências para funções não monotonicamente incrementais. Este trabalho estende a DIFT, visando incorporar um subconjunto das funções NMI em grafos dirigidos e mostrar sua aplicação no contexto da geração de superpixels. Outra aplicação que é apresentada para difundir a relevância das funções NMI é o algoritmo Bandeirantes para perseguição de bordas e rastreamento de curvas. / Image segmentation is a problem of great relevance in computer vision, in which an image is divided into relevant regions, such as to isolate an object of interest for a given application. Segmentation methods with monotonically incremental connectivity functions (MI) based on the Image Foresting Transform (IFT) have achieved great success in several contexts. In interactive segmentation of images, in which the user is allowed to specify the desired object, new seeds can be added and/or removed to correct the labeling until achieving the expected segmentation. This process generates a sequence of IFTs that can be calculated more efficiently by the Differential Image Foresting Trans- form (DIFT). Recently, non-monotonically incremental connectivity functions (NMI) have been used successfully in the IFT framework in the context of image segmentation, allowing the incorporation of shape, boundary polarity, and connectivity constraints, in order to customize the segmentation for a given target object. Non-monotonically incremental functions were also successfully exploited in the generation of superpixels, via sequences of IFT executions. In this work, we present a study of the Differential Image Foresting Transform in the case of NMI functions. Our research indicates that the original DIFT algorithm presents a series of inconsistencies for non-monotonically incremental functions. This work extends the DIFT algorithm to NMI functions in directed graphs, and shows its application in the context of the generation of superpixels. Another application that is presented to spread the relevance of NMI functions is the Bandeirantes algorithm for curve tracing and boundary tracking.

Le traitement d’images est un axe de recherche très actif depuis des années. L’interprétation des images constitue une de ses branches les plus importantes de par ses applications socio-économiques et scientifiques. Cependant cette interprétation, comme la plupart des processus de traitements d’images, nécessite une phase de segmentation pour délimiter les régions à analyser. En fait l’interprétation est un traitement qui permet de donner un sens aux régions détectées par la phase de segmentation. Ainsi, la phase d’interprétation ne pourra analyser que les régions détectées lors de la segmentation. Bien que l’objectif de l’interprétation automatique soit d’avoir le même résultat qu’une interprétation humaine, la logique des techniques classiques de ce domaine ne marie pas celle de l’interprétation humaine. La majorité des approches classiques d’interprétation d’images séparent la phase de segmentation et celle de l’interprétation. Les images sont d’abord segmentées puis les régions détectées sont interprétées. En plus, au niveau de la segmentation les techniques classiques parcourent les images de manière séquentielle, dans l’ordre de stockage des pixels. Ce parcours ne reflète pas nécessairement le parcours de l’expert humain lors de son exploration de l’image. En effet ce dernier commence le plus souvent par balayer l’image à la recherche d’éventuelles zones d’intérêts. Dans le cas échéant, il analyse les zones potentielles sous trois niveaux de vue pour essayer de reconnaitre de quel objet s’agit-il. Premièrement, il analyse la zone en se basant sur ses caractéristiques physiques. Ensuite il considère les zones avoisinantes de celle-ci et enfin il zoome sur toute l’image afin d’avoir une vue complète tout en considérant les informations locales à la zone et celles de ses voisines. Pendant son exploration, l’expert, en plus des informations directement obtenues sur les caractéristiques physiques de l’image, fait appel à plusieurs sources d’informations qu’il fusionne pour interpréter l’image. Ces sources peuvent inclure les connaissent acquises grâce à son expérience professionnelle, les contraintes existantes entre les objets de ce type d’images, etc. L’idée de l’approche présentée ici est que simuler l’activité visuelle de l’expert permettrait une meilleure compatibilité entre les résultats de l’interprétation et ceux de l’expert. Ainsi nous retenons de cette analyse trois aspects importants du processus d’interprétation d’image que nous allons modéliser dans l’approche proposée dans ce travail : 1. Le processus de segmentation n’est pas nécessairement séquentiel comme la plus part des techniques de segmentations qu’on rencontre, mais plutôt une suite de décisions pouvant remettre en cause leurs prédécesseurs. L’essentiel étant à la fin d’avoir la meilleure classification des régions. L’interprétation ne doit pas être limitée par la segmentation. 2. Le processus de caractérisation d’une zone d’intérêt n’est pas strictement monotone i.e. que l’expert peut aller d’une vue centrée sur la zone à vue plus large incluant ses voisines pour ensuite retourner vers la vue contenant uniquement la zone et vice-versa. 3. Lors de la décision plusieurs sources d’informations sont sollicitées et fusionnées pour une meilleure certitude. La modélisation proposée de ces trois niveaux met particulièrement l’accent sur les connaissances utilisées et le raisonnement qui mène à la segmentation des images. / Image processing has been a very active area of research for years. The interpretation of images is one of its most important branches because of its socio-economic and scientific applications. However, the interpretation, like most image processing processes, requires a segmentation phase to delimit the regions to be analyzed. In fact, interpretation is a process that gives meaning to the regions detected by the segmentation phase. Thus, the interpretation phase can only analyze the regions detected during the segmentation. Although the ultimate objective of automatic interpretation is to produce the same result as a human, the logic of classical techniques in this field does not marry that of human interpretation. Most conventional approaches to this task separate the segmentation phase from the interpretation phase. The images are first segmented and then the detected regions are interpreted. In addition, conventional techniques of segmentation scan images sequentially, in the order of pixels appearance. This way does not necessarily reflect the way of the expert during the image exploration. Indeed, a human usually starts by scanning the image for possible region of interest. When he finds a potential area, he analyzes it under three view points trying to recognize what object it is. First, he analyzes the area based on its physical characteristics. Then he considers the region's surrounding areas and finally he zooms in on the whole image in order to have a wider view while considering the information local to the region and those of its neighbors. In addition to information directly gathered from the physical characteristics of the image, the expert uses several sources of information that he merges to interpret the image. These sources include knowledge acquired through professional experience, existing constraints between objects from the images, and so on.The idea of the proposed approach, in this manuscript, is that simulating the visual activity of the expert would allow a better compatibility between the results of the interpretation and those ofthe expert. We retain from the analysis of the expert's behavior three important aspects of the image interpretation process that we will model in this work: 1. Unlike what most of the segmentation techniques suggest, the segmentation process is not necessarily sequential, but rather a series of decisions that each one may question the results of its predecessors. The main objective is to produce the best possible regions classification. 2. The process of characterizing an area of interest is not a one way process i.e. the expert can go from a local view restricted to the region of interest to a wider view of the area, including its neighbors and vice versa. 3. Several information sources are gathered and merged for a better certainty, during the decision of region characterisation. The proposed model of these three levels places particular emphasis on the knowledge used and the reasoning behind image segmentation.

Image segmentation refers to the process of partitioning an image into meaningful regions of interest (objects) by assigning distinct labels to their composing pixels. Images are usually composed of multiple objects with distinctive features, thus requiring distinct high-level priors for their appropriate modeling. In order to obtain a good segmentation result, the segmentation method must attend all the individual priors of each object, as well as capture their inclusion/exclusion relations. However, many existing classical approaches do not include any form of structural information together with different high-level priors for each object into a single energy optimization. Consequently, they may be inappropriate in this context. We propose a novel efficient seed-based method for the multiple object segmentation of images based on graphs, named Hierarchical Layered Oriented Image Foresting Transform (HLOIFT). It uses a tree of the relations between the image objects, being each object represented by a node. Each tree node may contain different individual high-level priors and defines a weighted digraph, named as layer. The layer graphs are then integrated into a hierarchical graph, considering the hierarchical relations of inclusion and exclusion. A single energy optimization is performed in the hierarchical layered weighted digraph leading to globally optimal results satisfying all the high-level priors. The experimental evaluations of HLOIFT and its extensions, on medical, natural and synthetic images, indicate promising results comparable to the state-of-the-art methods, but with lower computational complexity. Compared to hierarchical segmentation by the min cut/max-flow algorithm, our approach is less restrictive, leading to globally optimal results in more general scenarios, and has a better running time. / A segmentação de imagem refere-se ao processo de particionar uma imagem em regiões significativas de interesse (objetos), atribuindo rótulos distintos aos seus pixels de composição. As imagens geralmente são compostas de vários objetos com características distintas, exigindo, assim, restrições de alto nível distintas para a sua modelagem apropriada. Para obter um bom resultado de segmentação, o método de segmentação deve atender a todas as restrições individuais de cada objeto, bem como capturar suas relações de inclusão/ exclusão. No entanto, muitas abordagens clássicas existentes não incluem nenhuma forma de informação estrutural, juntamente com diferentes restrições de alto nível para cada objeto em uma única otimização de energia. Consequentemente, elas podem ser inapropriadas nesse contexto. Estamos propondo um novo método eficiente baseado em sementes para a segmentação de múltiplos objetos em imagens baseado em grafos, chamado Hierarchical Layered Oriented Image Foresting Transform (HLOIFT). Ele usa uma árvore das relações entre os objetos de imagem, sendo cada objeto representado por um nó. Cada nó da árvore pode conter diferentes restrições individuais de alto nível, que são usadas para definir um dígrafo ponderado, nomeado como camada. Os grafos das camadas são então integrados em um grafo hierárquico, considerando as relações hierárquicas de inclusão e exclusão. Uma otimização de energia única é realizada no dígrafo hierárquico em camadas, levando a resultados globalmente ótimos, satisfazendo todas as restrições de alto nível. As avaliações experimentais do HLOIFT e de suas extensões, em imagens médicas, naturais e sintéticas,indicam resultados promissores comparáveis aos métodos do estado-da-arte, mas com menor complexidade computacional. Comparada à segmentação hierárquica pelo algoritmo min-cut/max-flow, nossa abordagem é menos restritiva, levando a resultados globalmente ótimo sem cenários mais gerais e com melhor tempo de execução.

Un nombre incalculable de documents est imprimé, numérisé, faxé, photographié chaque jour. Ces documents sont hybrides : ils existent sous forme papier et numérique. De plus les documents numériques peuvent être consultés et modifiés simultanément dans de nombreux endroits. Avec la disponibilité des logiciels d’édition d’image, il est devenu très facile de modifier ou de falsifier un document. Cela crée un besoin croissant pour un système d’authentification capable de traiter ces documents hybrides. Les solutions actuelles reposent sur des processus d’authentification séparés pour les documents papiers et numériques. D’autres solutions reposent sur une vérification visuelle et offrent seulement une sécurité partielle. Dans d’autres cas elles nécessitent que les documents sensibles soient stockés à l’extérieur des locaux de l’entreprise et un accès au réseau au moment de la vérification. Afin de surmonter tous ces problèmes, nous proposons de créer un algorithme de hachage sémantique pour les images de documents. Cet algorithme de hachage devrait fournir une signature compacte pour toutes les informations visuellement significatives contenues dans le document. Ce condensé permettra la création de systèmes de sécurité hybrides pour sécuriser tout le document. Ceci peut être réalisé grâce à des algorithmes d’analyse du document. Cependant ceux-ci ont besoin d’être porté à un niveau de performance sans précédent, en particulier leur fiabilité qui dépend de leur stabilité. Après avoir défini le contexte de l’étude et ce qu’est un algorithme stable, nous nous sommes attachés à produire des algorithmes stables pour la description de la mise en page, la segmentation d’un document, la reconnaissance de caractères et la description des zones graphiques. / An innumerable number of documents is being printed, scanned, faxed, photographed every day. These documents are hybrid : they exist as both hard copies and digital copies. Moreover their digital copies can be viewed and modified simultaneously in many places. With the availability of image modification software, it has become very easy to modify or forge a document. This creates a rising need for an authentication scheme capable of handling these hybrid documents. Current solutions rely on separate authentication schemes for paper and digital documents. Other solutions rely on manual visual verification and offer only partial security or require that sensitive documents be stored outside the company’s premises and a network access at the verification time. In order to overcome all these issues we propose to create a semantic hashing algorithm for document images. This hashing algorithm should provide a compact digest for all the visually significant information contained in the document. This digest will allow current hybrid security systems to secure all the document. This can be achieved thanks to document analysis algorithms. However those need to be brought to an unprecedented level of performance, in particular for their reliability which depends on their stability. After defining the context of this study and what is a stable algorithm, we focused on producing stable algorithms for layout description, document segmentation, character recognition and describing the graphical parts of a document.

A segmentação de imagens é um problema muito importante em visão computacional, no qual uma imagem é dividida em regiões relevantes, tal como para isolar objetos de interesse de uma dada aplicação. Métodos de segmentação baseados na transformada imagem-floresta (IFT, Image Foresting Transform), com funções de conexidade monotonicamente incrementais (MI) têm alcançado um grande sucesso em vários contextos. Na segmentação interativa de imagens, na qual o usuário pode especificar o objeto desejado, novas sementes podem ser adicionadas e/ou removidas para corrigir a rotulação até conseguir a segmentação esperada. Este processo gera uma sequência de IFTs que podem ser calculadas de modo mais eficiente pela DIFT (Differential Image Foresting Transform). Recentemente, funções de conexidade não monotonicamente incrementais (NMI) têm sido usadas com sucesso no arcabouço da IFT no contexto de segmentação de imagens, permitindo incorporar informações de alto nível, tais como, restrições de forma, polaridade de borda e restrição de conexidade, a fim de customizar a segmentação para um dado objeto desejado. Funções não monotonicamente incrementais foram também exploradas com sucesso na geração de superpixels, via sequências de execuções da IFT. Neste trabalho, apresentamos um estudo sobre a Transformada Imagem-Floresta Diferencial no caso de funções NMI. Nossos estudos indicam que o algoritmo da DIFT original apresenta uma série de inconsistências para funções não monotonicamente incrementais. Este trabalho estende a DIFT, visando incorporar um subconjunto das funções NMI em grafos dirigidos e mostrar sua aplicação no contexto da geração de superpixels. Outra aplicação que é apresentada para difundir a relevância das funções NMI é o algoritmo Bandeirantes para perseguição de bordas e rastreamento de curvas. / Image segmentation is a problem of great relevance in computer vision, in which an image is divided into relevant regions, such as to isolate an object of interest for a given application. Segmentation methods with monotonically incremental connectivity functions (MI) based on the Image Foresting Transform (IFT) have achieved great success in several contexts. In interactive segmentation of images, in which the user is allowed to specify the desired object, new seeds can be added and/or removed to correct the labeling until achieving the expected segmentation. This process generates a sequence of IFTs that can be calculated more efficiently by the Differential Image Foresting Trans- form (DIFT). Recently, non-monotonically incremental connectivity functions (NMI) have been used successfully in the IFT framework in the context of image segmentation, allowing the incorporation of shape, boundary polarity, and connectivity constraints, in order to customize the segmentation for a given target object. Non-monotonically incremental functions were also successfully exploited in the generation of superpixels, via sequences of IFT executions. In this work, we present a study of the Differential Image Foresting Transform in the case of NMI functions. Our research indicates that the original DIFT algorithm presents a series of inconsistencies for non-monotonically incremental functions. This work extends the DIFT algorithm to NMI functions in directed graphs, and shows its application in the context of the generation of superpixels. Another application that is presented to spread the relevance of NMI functions is the Bandeirantes algorithm for curve tracing and boundary tracking.

Les travaux de cette thèse concernent la détection des changements dans des séries chronologiques de photographies de paysages prises depuis le sol. Ce contexte de comparaison d'images successives est celui que rencontrent les géographes de l'environnement qui ont recours aux observatoires photographiques du paysage. Ces outils d'analyse et d'aide à la décision sont des bases de données de photographies constituées selon une méthodologie stricte de rephotographie de la même scène, à des pas de temps réguliers. Le nombre de clichés est parfois très important, et l'analyse humaine fastidieuse et relativement imprécise, aussi un outil automatisant la comparaison de photos de paysage deux à deux pour mettre en évidence les changements serait une aide considérable dans l'exploitation des observatoires photographiques du paysage. Bien entendu, les variations dans l'éclairement, la saisonnalité, l'heure du jour, produisent fatalement des clichés entièrement différents à l'échelle du pixel. Notre objectif était donc de concevoir un système robuste face à ces changements mineurs, mais capable de détecter les changements pertinents de l'environnement. De nombreux travaux autour de la détection des changements ont été effectués pour des images provenant de satellites. Mais l'utilisation d'appareils photographiques numériques classiques depuis le sol pose des problèmes spécifiques comme la limitation du nombre de bandes spectrales et la forte variation de profondeur dans une même image qui induit des apparences différentes des mêmes catégories d'objets en fonction de leurs positions dans la scène. Dans un premier temps, nous avons exploré la voie de la détection automatique des changements. Nous avons proposé une méthode reposant sur le recalage et la sur-segmentation des images en superpixels. Ces derniers sont ensuite décrits par leur niveau de gris moyen ainsi que par leur texture au travers d'une représentation sous la forme d'histogrammes de textons. La distance de Mahalanobis entre ces descripteurs permet de comparer les superpixels correspondants entre deux images prises à des dates différentes. Nous avons évalué les performances de cette approche sur des images de l'observatoire photographique du paysage constitué lors de la construction de l'autoroute A89. Parmi les méthodes de segmentation utilisées pour produire les superpixels, les expérimentations que nous avons menées ont mis en évidence le bon comportement de la méthode de segmentation d'Achanta. La pertinence d'un changement étant fortement liée à l'application visée, nous avons exploré dans un second temps une piste faisant intervenir l'utilisateur. Nous avons proposé une méthode interactive de détection des changements reposant sur une phase d'apprentissage. Afin de détecter les changements entre deux images, l'utilisateur désigne, grâce à un outil de sélection, des échantillons constitués d'ensembles de pixels correspondant à des zones de changement et à des zones d'absence de changement. Chaque couple de pixels correspondants, c'est-à-dire situés au même endroit dans les deux images, est décrit par un vecteur de 16 valeurs principalement calculées à partir de l'image des dissemblances. Cette dernière est obtenue en mesurant, pour chaque couple de pixels correspondants, la dissemblance des niveaux de gris de leurs voisinages. Les échantillons désignés par l'utilisateur permettent de constituer des données d'apprentissage qui sont utilisées pour entraîner un classifieur. Parmi les méthodes de classification évaluées, les résultats expérimentaux montrent que les forêts d'arbres décisionnels donnent les meilleurs résultats sur les séries photographiques que nous avons utilisées. / This work deals with change detection from chronological series of photographs acquired from the ground. This context of consecutive images comparison is the one encountered in the field of integrated geography where photographic landscape observatories are widely used. These tools for analysis and decision-making consist of databases of photographic images obtained by strictly rephotographing the same scene at regular time intervals. With a large number of images, the human analysis is tedious and inaccurate. So a tool for automatically comparing pairs of landscape photographs in order to highlight changes would be a great help for exploiting photographic landscape observatories. Obviously, lighting variations, seasonality, time of day induce completely different images at the pixel level. Our goal is to design a system which would be robust to these insignificant changes and able to detect relevant changes of the scene. Numerous studies have been conducted on change detection from satellite images. But the utilization of classic digital cameras from the ground raise some specific problems like the limitation of the spectral band number and the strong variation of the depth in a same image which induces various appearance of the same object categories depending on their position in the scene. In the first part of our work, we investigate the track of automatic change detection. We propose a method lying on the registration and the over-segmentation of the images into superpixels. Then we describe each superpixel by its texture using texton histogram and its gray-level mean. A distance measure, such as Mahalanobis distance, allows to compare corresponding superpixels between two images acquired at different dates. We evaluate the performance of the proposed approach on images taken from the photographic landscape observatory produced during the construction of the French A89 highway. Among the image segmentation methods we have tested for superpixel extraction, our experiments show the relatively good behavior of Achanta segmentation method. The relevance of a change is strongly related to the intended application, we thus investigate a second track involving a user intervention. We propose an interactive change detection method based on a learning step. In order to detect changes between two images, the user designates with a selection tool some samples consisting of pixel sets in "changed" and "unchanged" areas. Each corresponding pixel pair, i.e., located at the same coordinates in the two images, is described by a 16-dimensional feature vector mainly calculated from the dissimilarity image. The latter is computed by measuring, for each corresponding pixel pair, the dissimilarity of the gray-levels of the neighbors of the two pixels. Samples selected by the user are used as learning data to train a classifier. Among the classification methods we have tried, experimental results indicate that random forests give the better results for the tested image series.

Modern automobiles and particularly those with off-road lineage possess subsystems that can be configured to better negotiate certain terrain types. Different terrain classes amount to different adherence (or surface grip) and compressibility properties that impact vehicle ma-noeuvrability and should therefore incur a tailored throttle response, suspension stiffness and so on. This thesis explores prospective terrain recognition for an anticipating terrain response driver assistance system. Recognition of terrain and road terrain is cast as a semantic segmen-tation task whereby forward driving images or point clouds are pre-segmented into atomic units and subsequently classified. Terrain classes are typically of amorphous spatial extent con-taining homogenous or granularly repetitive patterns. For this reason, colour and texture ap-pearance is the saliency of choice for monocular vision. In this work, colour, texture and sur-face saliency of atomic units are obtained with a bag-of-features approach. Five terrain classes are considered, namely grass, dirt, gravel, shrubs and tarmac. Since colour can be ambiguous among terrain classes such as dirt and gravel, several texture flavours are explored with scalar and structured output learning in a bid to devise an appropriate visual terrain saliency and predictor combination. Texture variants are obtained using local binary patters (LBP), filter responses (or textons) and dense key-point descriptors with daisy. Learning algorithms tested include support vector machine (SVM), random forest (RF) and logistic regression (LR) as scalar predictors while a conditional random field (CRF) is used for structured output learning. The latter encourages smooth labelling by incorporating the prior knowledge that neighbouring segments with similar saliency are likely segments of the same class. Once a suitable texture representation is devised the attention is shifted from monocular vision to stereo vision. Sur-face saliency from reconstructed point clouds can be used to enhance terrain recognition. Pre-vious superpixels span corresponding supervoxels in real world coordinates and two surface saliency variants are proposed and tested with all predictors: one using the height coordinates of point clouds and the other using fast point feature histograms (FPFH). Upon realisation that road recognition and terrain recognition can be assumed as equivalent problems in urban en-vironments, the top most accurate models consisting of CRFs are augmented with composi-tional high order pattern potentials (CHOPP). This leads to models that are able to strike a good balance between smooth local labelling and global road shape. For urban environments the label set is restricted to road and non-road (or equivalently tarmac and non-tarmac). Ex-periments are conducted using a proprietary terrain dataset and a public road evaluation da-taset.

Orientador: Nilton Nobuhiro Imai / Resumo: O objetivo desse doutorado é propor uma nova metodologia para identificar oito espécies arbóreas emergentes (i.e., que se sobressaem do dossel florestal), em diferentes idades e estágios de desenvolvimento e pertencentes à Mata Atlântica brasileira. Para tal, imagens hiperespectrais foram adquiridas em Julho/2017, em Junho/2018, e em Julho/2019 em um transecto localizado no fragmento florestal Ponte Branca, localizado a Oeste do Estado de São Paulo, onde a floresta é considerada estacional semidecidual e submontana. As imagens com resolução espacial de 10 cm foram adquiridas com câmara hiperespectral (500–900 nm) acoplada em veículo aéreo não tripulado (VANT ou UAV, do inglês Unmanned aerial vehicle) e, posteriormente corrigidas geometricamente e radiometricamente. Em seguida, as copas arbóreas individuais (ITCs, do inglês Individual tree crows) foram delineadas manualmente em cada conjunto de dados para serem utilizadas como referência para os experimentos. Dentre os experimentos realizados, destaca-se o uso do espectro normalizado para redução da variabilidade espectral intra-espécies, o uso da classificação baseada em regiões utilizando o algoritmo Random Forest e o uso de superpixexls para delineamento automático das ITCs em cada conjunto de imagens. Além disso, avaliou-se o uso dos superpixels multitemporais com diferentes atributos multitemporais (espectro normalizado, textura e índices de vegetação) e estruturais (derivados do modelo de altura das copas), sozinhos ou c... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The objective of this doctoral dissertation is to propose a new methodology to identify eight emergent tree species (i.e., that stood out from the canopy) belonging to highly diverse Brazilian Atlantic forest and with different ages and development stages. To achieve the objective, hyperspectral images were acquired in July/2017, June/208, and July/2019 in a transect area located in the western part of São Paulo State. The area is in Ponte Branca ecological station, where the forest is classified as submontane semideciduous seasonal with different stages of succession. Images with a spatial resolution of 10 cm were acquired with a hyperspectral camera (500–900 nm) onboard unmanned aerial vehicle (UAV) and geometrically and radiometrically post-processed. In sequence, the individual tree crowns (ITCs) were manually delineated in each dataset to be used as reference in the experiments. From the performed experiments, it is highlighted the use of mean normalized spectra to reduce the within-species spectral variability, the use of region-based classification with the Random Forest algorithm, and the use of superpixels to automatically delineate the ITCs in each dataset. Additionally, the multitemporal superpixels with different multitemporal features (normalized spectra, texture and vegetation indexes) and structural features derived from the canopy height model, combined or not, were assessed to the tree species classification. The best result was achieved merging normalized sp... (Complete abstract click electronic access below) / Doutor

INTRODUCTION: Despite the advances in radiotherapy treatment delivery, target volume delineation remains one of the greatest sources of error in the radiotherapy delivery process, which can lead to poor tumour control probability and impact clinical outcome. Contouring assessments are performed to ensure high quality of target volume definition in clinical trials but this can be subjective and labour-intensive. This project addresses the hypothesis that computational segmentation techniques, with a given prior, can be used to develop an image-based tumour delineation process for contour assessments. This thesis focuses on the exploration of the segmentation techniques to develop an automated method for generating reference delineations in the setting of advanced lung cancer. The novelty of this project is in the use of the initial clinician outline as a prior for image segmentation. METHODS: Automated segmentation processes were developed for stage II and III non-small cell lung cancer using the IDEAL-CRT clinical trial dataset. Marker-controlled watershed segmentation, two active contour approaches (edge- and region-based) and graph-cut applied on superpixels were explored. k-nearest neighbour (k-NN) classification of tumour from normal tissues based on texture features was also investigated. RESULTS: 63 cases were used for development and training. Segmentation and classification performance were evaluated on an independent test set of 16 cases. Edge-based active contour segmentation achieved highest Dice similarity coefficient of 0.80 ± 0.06, followed by graphcut at 0.76 ± 0.06, watershed at 0.72 ± 0.08 and region-based active contour at 0.71 ± 0.07, with mean computational times of 192 ± 102 sec, 834 ± 438 sec, 21 ± 5 sec and 45 ± 18 sec per case respectively. Errors in accuracy of irregularly shaped lesions and segmentation leakages at the mediastinum were observed. In the distinction of tumour and non-tumour regions, misclassification errors of 14.5% and 15.5% were achieved using 16- and 8-pixel regions of interest (ROIs) respectively. Higher misclassification errors of 24.7% and 26.9% for 16- and 8-pixel ROIs were obtained in the analysis of the tumour boundary. CONCLUSIONS: Conventional image-based segmentation techniques with the application of priors are useful in automatic segmentation of tumours, although further developments are required to improve their performance. Texture classification can be useful in distinguishing tumour from non-tumour tissue, but the segmentation task at the tumour boundary is more difficult. Future work with deep-learning segmentation approaches need to be explored.

Είναι ενδιαφέρουσα αλλά συνάμα δύσκολη η ανάλυση ιατρικών εικόνων, επειδή υπάρχουν πολύ μικρές διακυμάνσεις και μεγάλος όγκος δεδομένων για επεξεργασία. Είναι αρκετά δύσκολο να αναπτυχθεί ένα αυτοματοποιημένο σύστημα αναγνώρισης, το οποίο θα μπορούσε να επεξεργάζεται μεγάλο όγκο πληροφοριών των ασθενών και να παρέχει μια σωστή εκτίμηση. Στην ιατρική, η συμβατική διαγνωστική μέθοδος για εικόνες MR γονάτου για αναγνώριση ανωμαλιών, είναι από την επίβλεψη έμπειρων ιατρών. Η τεχνική της ασαφούς λογικής είναι πιο ακριβής, αλλά αυτό εξαρτάται πλήρως από τη γνώση των εμπειρογνωμόνων, η οποία μπορεί να μην είναι πάντα διαθέσιμη. Στη παρούσα εργασία, τμηματοποιούμε την MR εικόνα του γονάτου με την τεχνική Mean Shift, αναγνωρίζουμε τα κύρια μέρη με τη βοήθεια των ΗΜRF και τέλος εκπαιδεύουμε ταξινομητή ANFIS. Η απόδοση του ταξινομητή ANFIS αξιολογήθηκε όσον αφορά την απόδοση της εκαπαίδευσης και της ακρίβειας ταξινόμησης. Επιβεβαιώθηκε ότι ο ταξινομητής είχε μεγάλη ακρίβεια στην ανίχνευση ανωμαλιών στις ακτινογραφίες Στην εργασία αυτή περιγράφεται η προτεινόμενη στρατηγική για την διάγνωση ανωμαλιών στις εικόνες μαγνητικής τομογραφίας γόνατος. / It is a challenging task to analyze medical images because there are very minute variations & larger data set for analysis. It is a quite difficult to develop an automated recognition system which could process on a large information of patient and provide a correct estimation. The conventional method in medicine for knee MR images classification and diseases detection is by human inspection. Fuzzy logic technique is more accurate but it fully depends on expert knowledge, which may not always available. Here we extract the feature using Mean Shift segmentation and region recognition with HMRF and after that training using the ANFIS tool. The performance of the ANFIS classifier was evaluated in terms of training performance and classification accuracy. Here the result confirmed that the proposed ANFIS classifier with high accuracy in detecting the knee diseases. This work describes the proposed strategy to medical image classification of patient’s MRI scan images of the knee.