This thesis focuses on implementation of an algorithm for retinal vessel segmentation in high resolution retinal images.A neural network with two hidden layers was used as the method. A total of 7 features were obtained from matched filtering based on vessel thickness, texture analysis and individual pixels brightness. Within the thesis, the whole database was manually annotated for the implementation of the algorithm and the results. The achieved mean sensitivity reached 80%, specificity 70% and Dice coefficient is 59%.

The goal of this work was to design fast and simple methods for processing point-cloud-data of urban areas for virtual reality applications. For the visualization of methods, we developed a simple renderer written in C++ and HLSL. The renderer is based on DirectX 11. For point-cloud processing, we designed a method based on height-histograms for filtering ground points out of point cloud. We also proposed a parallel method for point cloud segmentation based on the region growing algorithm. The individual segments are then tested by simple rules to check if it is or it is not corresponding to a predefined object.

La gradation de lames de biopsie fournit des informations pronostiques essentielles pour le diagnostic et le traitement. La détection et le comptage manuel des mitoses est un travail fastidieux, sujet à des variations inter-et intra- observateur considérables. L'objectif principal de cette thèse de doctorat est le développement d'un système capable de fournir une détection des mitoses sur des images provenant de différents types de scanners rapides automatiques, ainsi que d'un microscope multispectral. L'évaluation des différents systèmes proposés est effectuée dans le cadre du projet MICO (MIcroscopie COgnitive, projet ANR TecSan piloté par notre équipe). Dans ce contexte, les systèmes proposés ont été testés sur les données du benchmark MITOS. En ce qui concerne les images couleur, notre système s'est ainsi classé en deuxième position de ce concours international, selon la valeur du critère F-mesure. Par ailleurs, notre système de détection de mitoses sur images multispectrales surpasse largement les meilleurs résultats obtenus durant le concours. / Digital pathology represents one of the major and challenging evolutions in modernmedicine. Pathological exams constitute not only the gold standard in most of medicalprotocols, but also play a critical and legal role in the diagnosis process. Diagnosing adisease after manually analyzing numerous biopsy slides represents a labor-intensive workfor pathologists. Thanks to the recent advances in digital histopathology, the recognitionof histological tissue patterns in a high-content Whole Slide Image (WSI) has the potentialto provide valuable assistance to the pathologist in his daily practice. Histopathologicalclassification and grading of biopsy samples provide valuable prognostic information thatcould be used for diagnosis and treatment support. Nottingham grading system is thestandard for breast cancer grading. It combines three criteria, namely tubule formation(also referenced as glandular architecture), nuclear atypia and mitosis count. Manualdetection and counting of mitosis is tedious and subject to considerable inter- and intrareadervariations. The main goal of this dissertation is the development of a framework ableto provide detection of mitosis on different types of scanners and multispectral microscope.The main contributions of this work are eight fold. First, we present a comprehensivereview on state-of-the-art methodologies in nuclei detection, segmentation and classificationrestricted to two widely available types of image modalities: H&E (HematoxylinEosin) and IHC (Immunohistochemical). Second, we analyse the statistical and morphologicalinformation concerning mitotic cells on different color channels of various colormodels that improve the mitosis detection in color datasets (Aperio and Hamamatsu scanners).Third, we study oversampling methods to increase the number of instances of theminority class (mitosis) by interpolating between several minority class examples that lietogether, which make classification more robust. Fourth, we propose three different methodsfor spectral bands selection including relative spectral absorption of different tissuecomponents, spectral absorption of H&E stains and mRMR (minimum Redundancy MaximumRelevance) technique. Fifth, we compute multispectral spatial features containingpixel, texture and morphological information on selected spectral bands, which leveragediscriminant information for mitosis classification on multispectral dataset. Sixth, we performa comprehensive study on region and patch based features for mitosis classification.Seven, we perform an extensive investigation of classifiers and inference of the best one formitosis classification. Eight, we propose an efficient and generic strategy to explore largeimages like WSI by combining computational geometry tools with a local signal measureof relevance in a dynamic sampling framework.The evaluation of these frameworks is done in MICO (COgnitive MIcroscopy, ANRTecSan project) platform prototyping initiative. We thus tested our proposed frameworks on MITOS international contest dataset initiated by this project. For the color framework,we manage to rank second during the contest. Furthermore, our multispectral frameworkoutperforms significantly the top methods presented during the contest. Finally, ourframeworks allow us reaching the same level of accuracy in mitosis detection on brightlightas multispectral datasets, a promising result on the way to clinical evaluation and routine.

Les plus grandes villes au monde disposent de plans 2D très détaillés des rues et des espaces publics. Ces plans contiennent des informations relatives aux routes, trottoirs, façades et objets urbains tels que, entre autres, les lampadaires, les panneaux de signalisation, les poteaux, et les arbres.De nos jours, certaines autorités locales, agences nationales de cartographie et sociétés privées commencent à adjoindre à leurs cartes de villes des informations en 3D, des choix de navigation et d'accessibilité.En comparaison des premiers systèmes de scanning en 3D d'il y a 30 ans, les scanners laser actuels sont moins chers, plus rapides et fournissent des nuages de points 3D plus précis et plus denses.L'analyse de ces données est difficile et laborieuse, et les méthodes semi-automatiques actuelles risquent de ne pas être suffisamment précises ni robustes. C'est en ce sens que des méthodes automatiques pour l'analyse urbaine sémantique en 3D sont nécessaires.Cette thèse constitue une contribution au domaine de l'analyse sémantique de nuages de points en 3D dans le cadre d'un environnement urbain.Nos méthodes sont basées sur les images d'élévation et elles illustrent l'efficacité de la morphologie mathématique pour développer une chaîne complète de traitement en 3D, incluant 6 étapes principales:i)~filtrage et pré-traitement;ii)~segmentation du sol et analyse d'accessibilité;iii)~segmentation des façades;iv)~détection d'objets;v)~segmentation d'objets;vi)~classification d'objets.De plus, nous avons travaillé sur l'intégration de nos résultats dans une chaîne de production à grande échelle.Ainsi, ceux-ci ont été incorporés en tant que ``shapefiles'' aux Systèmes d'Information Géographique et exportés en tant que nuages de points 3D pour la visualisation et la modélisation.Nos méthodes ont été testées d'un point de vue qualitatif et quantitatif sur plusieurs bases de données issues de l'état de l'art et du projet TerraMobilita.Nos résultats ont montré que nos méthodes s'avèrent précises, rapides et surpassent les travaux décrits par la littérature sur ces mêmes bases.Dans la conclusion, nous abordons également les perspectives de développement futur. / Most important cities in the world have very detailed 2D urban plans of streets and public spaces.These plans contain information about roads, sidewalks, facades and urban objects such as lampposts, traffic signs, bollards, trees, among others.Nowadays, several local authorities, national mapping agencies and private companies have began to consider justifiable including 3D information, navigation options and accessibility issues into urban maps.Compared to the first 3D scanning systems 30 years ago, current laser scanners are cheaper, faster and provide more accurate and denser 3D point clouds.Urban analysis from these data is difficult and tedious, and existing semi-automatic methods may not be sufficiently precise nor robust.In that sense, automatic methods for 3D urban semantic analysis are required.This thesis contributes to the field of semantic analysis of 3D point clouds from urban environments.Our methods are based on elevation images and illustrate how mathematical morphology can be exploited to develop a complete 3D processing chain including six main steps:i)~filtering and preprocessing;ii)~ground segmentation and accessibility analysis;iii)~facade segmentation,iv)~object detection;v)~object segmentation;and, vi)~object classification.Additionally, we have worked on the integration of our results into a large-scale production chain. In that sense, our results have been exported as 3D point clouds for visualization and modeling purposes and integrated as shapefiles into Geographical Information Systems (GIS).Our methods have been qualitative and quantitative tested in several databases from the state of the art and from TerraMobilita project.Our results show that our methods are accurate, fast and outperform other works reported in the literature on the same databases.Conclusions and perspectives for future work are discussed as well.

Classification of the agricultural fields using remote sensing images is one of the most popular methods used for crop mapping. Most recent classification techniques are based on per-field approach that works as assigning a crop label for each field. Commonly, the spatial vector data is used for the boundaries of the fields for applying the classification within them. However, crop variation within the fields is a very common problem. In this case, the existing field boundaries may be insufficient for performing the field-based classification and therefore, image segmentation is needed to be employed to detect these homogeneous segments within the fields. This study proposed a field-based approach to segment the crop fields in an image within the integrated environment of Geographic Information System (GIS) and Remote Sensing. In this method, each field is processed separately and the segments within each field are detected. First, an edge detection is applied to the images, and the detected edges are vectorized to generate the straight line segments. Next, these line segments are correlated with the existing field boundaries using the perceptual grouping techniques to form the closed regions in the image. The closed regions represent the segments each of which contain a distinct crop type. To implement the proposed methodology, a software was developed. The implementation was carried out using the 10 meter spatial resolution SPOT 5 and the 20 meter spatial resolution SPOT 4 satellite images covering a part of Karacabey Plain, Turkey. The evaluations of the obtained results are presented using different band combinations of the images.

This work focuses on the topic of machine learning, specifically implementation of a program for automated classification using deep learning. This work compares different trainable models of neural networks and describes practical solutions encountered during their implementation.

This thesis deals with design and implementation of an algorithm for segmentation of ribs from thoracic CT data. For the segmentation method of rib centerlines detection is chosen. The first step of this approach is to extract the centerlines of all the bones located in the scan. These centerlines are divided into short primitives, which are subsequently classified into couple of categories, depending on whether they represent the centerline of the rib. Subsequently, the centrelines of ribs are used as the seed points of the region growing algorithm in three-dimensional space, which realizes the final segmentation of the ribs. Within the work, a database of 10 CT scans was manually annotated, which was subsequently used to validate a performance of the proposed segmentation approach. The achieved success rate of primitive classification is 96,7 %, the success rate of rib segmentation (Dice coefficient) is 86,8 %.

Registration of a group of images generally only gives a pointwise, dense correspondence defined over the whole image plane or volume, without having any specific description of any common structure that exists in every image. Furthermore, identifying tissue classes and structures that are significant across the group is often required for analysis, as well as the correspondence. The overall aim is instead to perform registration, segmentation, and modelling simultaneously, so that the registration can assist the segmentation, and vice versa. However, structural information does play a role in conventional registration, in that if the registration is successful, it would be expected structures to be aligned to some extent. Hence, we perform initial experiments to investigate whether there is explicit structural information present in the shape of the registration objective function about the optimum. We perturbed one image locally with a diffeomorphism, and found interesting structure in the shape of the quality of fit function. Then, we proceed to add explicit structural information into registration framework, using various types of structural information derived from the original intensity images. For the case of MR brain images, we augment each intensity image with its own set of tissue fraction images, plus intensity gradient images, which form an image ensemble for each example. Then, we perform groupwise registration by using these ensembles of images. We apply the method to four different real-world datasets, for which ground-truth annotation is available. It is shown that the method can give a greater than 25% improvement on the three difficult datasets, when compared to using intensity-based registration alone. On the easier dataset, it improves upon intensity-based registration, and achieves results comparable with the previous method.

Immunohistological (IHC) stained images occupy a fundamental role in the pathologist's diagnosis and monitoring of cancer development. The manual process of monitoring such images is a subjective, time consuming process that typically relies on the visual ability and experience level of the pathologist. A novel and comprehensive system for the automated quantification of antibody inside stained cell nuclei in immunohistochemistry images is proposed and demonstrated in this research. The system is based on a cellular level approach, where each nucleus is individually analyzed to observe the effects of protein antibodies inside the nuclei. The system provides three main quantitative descriptions of stained nuclei. The first quantitative measurement automatically generates the total number of cell nuclei in an image. The second measure classifies the positive and negative stained nuclei based on the nuclei colour, morphological and textural features. Such features are extracted directly from each nucleus to provide discriminative characteristics of different stained nuclei. The output generated from the first and second quantitative measures are used collectively to calculate the percentage of positive nuclei (PS). The third measure proposes a novel automated method for determining the staining intensity level of positive nuclei or what is known as the intensity score (IS). The minor intensity features are observed and used to classify low, intermediate and high stained positive nuclei. Statistical methods were applied throughout the research to validate the system results against the ground truth pathology data. Experimental results demonstrate the effectiveness of the proposed approach and provide high accuracy when compared to the ground truth pathology data.