Data extraction from engraved text is discussed rarely, and nothing in the open literature discusses data extraction from cemetery headstones. Headstone images present unique challenges such as engraved or embossed characters (causing inner-character shadows), low contrast with the background, and significant noise due to inconsistent stone texture and weathering. Current systems for extracting text from outdoor environments (billboards, signs, etc.) make assumptions (i.e. clean and/or consistently-textured background and text) that fail when applied to the domain of engraved text. Additionally, the ability to extract the data found on headstones is of great historical value. This thesis describes a novel and efficient feature-based text zoning and segmentation method for the extraction of noisy text from a highly textured engraved medium. Additionally, the usefulness of constraining a problem to a specific domain is demonstrated. The transcriptions of images zoned and segmented through the proposed system result in a precision of 55% compared to 1% precision without zoning, a 62% recall compared to 39%, an F-measure of 58% compared to 2%, and an error rate of 77% compared to 8303%.

We present a new technique for creating an editable vector graphic from an object in a raster image. Object selection is performed interactively in subsecond time by calling graph cut with each mouse movement. A renderable mesh is then computed automatically for the selected object and each of its (sub)objects by (1) generating a coarse object mesh; (2) performing recursive graph cut segmentation and hierarchical ordering of subobjects; (3) applying error-driven mesh refinement to each (sub)object. The result is a fully layered object hierarchy that facilitates object-level editing without leaving holes. Object-based vectorization compares favorably with current approaches in the representation and rendering quality. Object-based vectorization and complex editing tasks are performed in a few 10s of seconds.

A segmentação de imagens consiste em dividir uma imagem em regiões ou objetos que a compõem, como, por exemplo, para isolar os pixels de um objeto alvo de uma dada aplicação. Em segmentação de imagens médicas, o objeto de interesse comumente apresenta transições em suas bordas predominantemente do tipo claro para escuro ou escuro para claro. Métodos tradicionais por região, como a conexidade fuzzy relativa (RFC - Relative Fuzzy Connectedness), não distinguem bem entre essas bordas similares com orientações opostas. A especificação da polaridade de contorno pode ajudar a amenizar esse problema, o que requer uma formulação matemática em grafos dirigidos. Uma discussão sobre como incorporar essa propriedade no arcabouço do RFC é apresentada neste trabalho. Uma prova teórica da otimalidade do novo algoritmo, chamado conexidade fuzzy relativa com orientação (ORFC - Oriented Relative Fuzzy Connectedness), em termos de uma função de energia em grafos dirigidos sujeita as restrições de sementes é apresentada, bem como a sua apli- cação em poderosos métodos híbridos de segmentação. O método híbrido proposto ORFC &Graph Cut preserva a robustez do ORFC em relação à escolha de sementes, evitando o problema do viés de encolhimento do método de Corte em Grafo (GC - Graph Cut), e mantém o forte controle do GC no delineamento de contornos de bordas irregulares da imagem. Os métodos propostos são avaliados usando imagens médicas de ressonáncia magnética (RM) e tomografia computadorizada (TC) do cérebro humano e de estudos torácicos. / Image segmentation consists of dividing an image into its composing regions or objects, for example, to isolate the pixels of a target object of a given application. In segmentation of medical images, the object of interest commonly presents transitions at its border predominantly from bright to dark or dark to bright. Traditional region-based methods of image segmentation, such as Relative Fuzzy Connectedness (RFC), do not distinguish well between similar boundaries with opposite orientations. The specification of the boundary polarity can help to alleviate this problem but this requires a mathematical formulation on directed graphs. A discussion on how to incorporate this property in the RFC framework is presented in this work. A theoretical proof of the optimality of the new algorithm, called Oriented Relative Fuzzy Connectedness (ORFC), in terms of an energy function on directed graphs subject to seed constraints is presented, and its application in powerful hybrid segmentation methods. The hybrid method proposed ORFC&Graph Cut preserves the robustness of ORFC respect to the seed choice, avoiding the shrinking problem of Graph Cut (GC), and keeps the strong control of the GC in the contour delination of irregular image boundaries. The proposed methods are evaluated using magnetic resonance medical imaging (MR) and computed tomography (CT) of the human brain and thoracic studies.

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.

Retinal imaging enables the visualization of a portion of the human microvasculature in-vivo and non-invasively. The scanning laser ophthalmoscope (SLO), provides images characterized by an ultra-wide field of view (UWFoV) covering approximately 180-200º in a single scan, minimizing the discomfort for the subject. The microvasculature visible in retinal images and its changes have been vastly investigated as candidate biomarkers for different types of systemic conditions like cardiovascular disease (CVD), which currently remains the main cause of death in Europe. For the CARMEN study, UWFoV SLO images were acquired from more than 1,000 people who were recruited from two studies, TASCFORCE and SCOT-HEART, focused on CVD. This thesis presents an automated system for SLO image processing and computation of candidate biomarkers to be associated with cardiovascular risk and MRI imaging data. A vessel segmentation technique was developed by making use of a bank of multi-scale matched filters and a neural network classifier. The technique was devised to minimize errors in vessel width estimation, in order to ensure the reliability of width measures obtained from the vessel maps. After a step of refinement of the centrelines, a multi-level classification technique was deployed to label all vessel segments as arterioles or venules. The method exploited a set of pixel-level features for local classification and a novel formulation for a graph cut approach to partition consistently the retinal vasculature that was modelled as an undirected graph. Once all the vessels were labelled, a tree representation was adopted for each vessel and its branches to fully automate the process of biomarker extraction. Finally, a set of 75 retinal parameters, including information provided by the periphery of the retina, was created for each image and used for the biomarker investigation.

Title: Interactive Processing of Volumetric Data Author: Jan Kolomazník Department: Department of Software and Computer Science Education Supervisor: RNDr. Josef Pelikán, Department of Software and Computer Science Education Abstract: Interactive visualization and segmentation of volumetric data are quite lim- ited due to the increased complexity of the task and size of the input data in comparison to two-dimensional processing. A special interactive segmentation workflow is presented, based on minimal graph-cut search. The overall execution time was lowered by implementing all the computational steps on GPU, which required a design of massively parallel algorithms (using thousands of threads). To lower the computational burden even further the graph is constructed over the image subregions com- puted by parallel watershed transformation. As a suitable formalism for a range of massively parallel algorithms was chosen cellular automata. A set of cellular automata extensions was defined, which allows efficient mapping and computation on GPU. Several variants of parallel watershed transformation are then defined in the form of cellular automaton. A novel form of 2D transfer function was presented, to improve direct volume visualization of the input data, suited for discriminating image features by their shape and...

Dans le contexte général des thérapies minimalement invasives, les travaux de cette thèse portent sur le planning d'une thérapie interstitielle de tumeurs du foie par ultrasons haute intensité. Dans un premier temps, une caractérisation des structures anatomiques hépatiques à partir de données scanner X est proposée selon deux méthodes de segmentation basée sur le graph cut : l'une semi-interactive et rapide pour extraire le foie et les éventuelles tumeurs ; et l'autre automatique et spécifique à la segmentation de la vascularisation hépatique par l'introduction d'un a priori local de forme estimé à partir de moments géométriques 3D. La seconde partie de cette étude est consacrée à la modélisation des effets de la thérapie sur les tissus. Le modèle proposé offre la possibilité de simuler différents types de sonde composée d'une matrice d'éléments contrôlables en phase et intensité. La description de la vascularisation locale dans le milieu peut également être intégrée dans le modèle. Les travaux et résultats obtenus portent sur trois aspects et/ou applications de ce modèle : 1) une méthode pour accélérer la résolution de la BHTE sous certaines hypothèses, 2) des résultats préliminaires de modélisation d'une sonde 64 éléments à focalisation dynamique et 3) le design géométrique d'une sonde endocavitaire 256 éléments.

In this thesis, we would like to present a new way to enhance the “depth map” image which is called as the fusion of depth images. The goal of our thesis is to try to enhance the “depth images” through a fusion of different classification methods. For that, we will use three similar but different methodologies, the Graph-Cut, Super-Pixel and Principal Component Analysis algorithms to solve the enhancement and output of our result. After that, we will compare the effect of the enhancement of our result with the original depth images. This result indicates the effectiveness of our methodology. / Room 401, No.56, Lane 21, Yin Gao Road, Shanghai, China

This thesis deals with the application of graph theory in image segmentation. There are specifically presented method utilizing graph cuts and extensions of this method. In the first chapter thera are initially explained basics of graph theory that are essential for understanding of the presented method. It is described in the second chapter, including its extensions that use shape priors. In the third chapter there is presented solution which is used for vertebrae lesion segmentation in the CT data sets. Final function is implemented into the program but it can be used also separately. Success rate is described using sensitivity and specificity in the last chapter, there are also examples of results.

Submitted by Marlene Santos (marlene.bc.ufg@gmail.com) on 2017-09-04T16:56:52Z No. of bitstreams: 2 Dissertação - Lais Cândido Rodrigues da Silva Lopes - 2017.pdf: 3651786 bytes, checksum: 544801154cf6cd32456e6887eaa09b85 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2017-09-15T13:52:55Z (GMT) No. of bitstreams: 2 Dissertação - Lais Cândido Rodrigues da Silva Lopes - 2017.pdf: 3651786 bytes, checksum: 544801154cf6cd32456e6887eaa09b85 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2017-09-15T13:52:55Z (GMT). No. of bitstreams: 2 Dissertação - Lais Cândido Rodrigues da Silva Lopes - 2017.pdf: 3651786 bytes, checksum: 544801154cf6cd32456e6887eaa09b85 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2017-08-03 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / The capture of images by multiple positions allows to recover the three-dimensional information of the environment applying the knowledge about the geometry of the cameras and the correspondences between the points of the images. The correspondence of characteristics in images is the task of relating regions of different images to the same point of interest, being considered a problem of difficult solution, since it suffers with ambiguities, occlusions, variation of illumination, besides local distortions. For having so many challenges, this subject is one of the most investigated in the field of computer vision cite Scharstein2001. The present dissertation aims to generate dense disparity maps, using graph cutting, from search spaces constructed with matching metrics based on laws of the Gestalt theory. A hybrid approach was developed, consisting of a local algorithm to construct the image disparity space (EDI), and a global algorithm used to optimize the disparities. The results were maps of disparities close to the expected maps ( textit groundtruth). It was also perceived the best performance of the methodology proposed in relation to the separate methods that compose it. / A captura de imagens por múltiplas posições permite recuperar a informação tridimensional do ambiente aplicando o conhecimento sobre a geometria das câmeras e as correspondências entre os pontos das imagens. A correspondência de características em imagens é a tarefa de relacionar regiões de imagens diferentes a um mesmo ponto de interesse, sendo considerado um problema de difícil solução, uma vez que, sofre com ambiguidades, oclusões, variação de iluminação, além de distorções locais. Por contar com tantos desafios, este tema é um dos mais investigados na área de visão computacional [Scharstein e Szeliski 2002]. A presente dissertação tem por objetivo gerar mapas de disparidade densos, usando corte de grafos, a partir de espaços de busca construídos com métricas de correspondência baseadas em leis da teoria Gestalt. Foi desenvolvida uma abordagem híbrida, composta de um algoritmo local para construir o espaço de disparidades da imagem (EDI), e um algoritmo global utilizado para otimizar as disparidades. Os resultados foram mapas de disparidades próximos dos mapas esperados (ground-truth). Percebeu-se a melhor performance da metodologia proposta em relação aos métodos em separado que a compõe.