This master’s thesis is aimed at registration of frames of retinal fundus video using phase corre- lation. An introduction describes general research in topic of retinal fundus, eye movements, diff erent approaches for image registration, phase correlation and examples of phase corre- lation applications. The second, practical part of master’s thesis, is dedicated to description of the proposed algorithm for registration of frames of retinal fundus video. The description of the proposed algorithm is divided into three parts. First two parts describe how frames of retinal fundus video are rated in terms of suitability for registration. Third part describes image registration algorithm itself. In conclusion, the accuracy of algorithm and computational time are evaluated.

Point set registration is a well-researched yet still not a very exploited area in computer vision. As the field of machine learning grows, the possibilities of application expand. This thesis investigates the possibility to expand an already implemented probabilistic machine learning approach to point set registration to more complex, larger datasets gathered in a forest environment. The system used as a starting point was created by Järemo Lawin et. al. [10]. The aim of the thesis was to investigate the possibility to register the forest data with the existing system, without ground-truth poses, with different optimizers, and to implement a SLAM pipeline. Also, older methods were used as a benchmark for evaluation, more specifically iterative closest point(ICP) and fast global registration(FGR).To enable the gathered data to be processed by the registration algorithms, preprocessing was required. Transforming the data points from the coordinate system of the sensor to world relative coordinates via LiDAR base coordinates. Subsequently, the registration was performed with different approaches. Both the KITTI odometry dataset, which RLLReg originally was evaluated with[10], and the gathered forest data were used. Data augmentation was utilized to enable ground-truth-independent training and to increase diversity in the data. In addition, the registration results were used to create a SLAM-pipeline, enabling mapping and localization in the scanned areas. The results showed great potential for using RLLReg to register forest scenes compared to other, older, approaches. Especially, the lack of ground-truth was manageable using data augmentation to create training data. Moreover, there was no evidence that AdaBound improves the system when replacing the Adam-optimizer. Finally, forest models with sensor paths plotted were generated with decent results. However, a potential for post-processing with further refinement is possible. Nevertheless, the possibility of point set registration and LiDAR-SLAM using machine learning has been confirmed.

In the first part of my master's thesis the network architecture of IMS (IP Multimedia Subsystem) is presented. The database of subscribers HSS (Home Subscriber Server) and SLF (Subscription Locator Function), as well as a SIP CSCF servers (Call Session Control Functions) process a SIP signalization and an AS application server performing services, etc. I focus on the registration of subscribers in the IMS network with a list of transmitted messages and description of each interface that is used in this network. The most important interfaces, which I described here, are Gm, Mw, Cx, Dx and Sh. Then I focused on security in IMS problems, which are divided into categories of access security and network security. After that is the implementation of IMS network in an open source Open IMS Core System considered under the Linux operating system. Here is the problem description from the actual system installation, through the configuration of all necessary elements of the network to the communication party itself. The communication analysis in the initial registration process and in subsequent communications is described. Finally I created laboratory exercises with a focus on the Open IMS Core System, where students learn about architecture and principle of networks based on IMS technology operation, with individual elements necessary for the operation of the network and their configuration. Students also test simple captured traffic analysis.

This master thesis deals with retinal image registration using phase correlation technique. There are described properties of retinal images and modality of scanning. A geometrical transformation encompasing scale, rotation and translation between two retinal images is considered and the whole registration framework is described. There are used retinal images from fundus camera and scanning laser ophthalmoscope (SLO). In this thesis is described corresponding bifurcations detection using phase correlation and registration using second-order polynomial transformation. The results are subjectively and objectively verificated.

The aim of this thesis is to familiarize with the issue of registration of medical image sequences. The main objective was to focus on the method of feature tracking in the image and various options of its implementation. The theoretical part describes various methods for detection of feature points and future point matching methods. In the practical part these methods were implemented in Matlab programming environment and a simple graphical user interface was created.

Tato práce se zabývá použitím fázové korelace k určení vzájemné rotace, změny měřítka a posunu mezi digitálními obrazy. Fázová korelace je založena na Fourierově transformaci, proto je popsána Fourierova transformace funkcí definovaných na R^2 i diskrétní Fourierova transformace funkcí definovaných na konečném počtu bodů {0, 1, ... , N-1}^2, kde N je přirozené číslo. Dále je pozornost věnována modifikacím fázové korelace, díky nimž metoda umožňuje nalezení parametrů podobnostní transformace i mezi obrazy, které mají vysoký dynamický rozsah a slabě patrné struktury, obsahují aditivní nebo impulzní šum a jsou pořízeny pomocí různých snímačů a optických soustav. Obsahem práce jsou i modifikace metody pro snímky sluneční koróny pořízené během úplných zatmění Slunce, což patří mezi nejobtížnější úlohy registrace obrazů.

This work is focused on creating 3D maps of indoor enviroment using Microsoft Kinect sensor. The first part shows the description of Microsoft Kinect sensor, the methods for acquisition and processing of depth data and their registration using different algorithms. The second part shows application of algorithms for map registration and final 3D maps of indoor enviroment.

Tato diplomová práce popisuje metodu využití lícování obrazů pro korekci geometrické distorze obrazů pravidelných krystalických struktur získaných z transmisního elektronového mikroskopu (TEM). Cílem této práce je vytvořit algoritmus v Matlabu, který dokáže tyto vady eliminovat nalezením prostorové transformace, která nalícuje zkreslený obraz na jeho modelovou mřížku. Transformace je hledána s využitím optimalizačních metod, které optimalizují zvolenou kriteriální funkci.

L’évaluation de la gravité et la surveillance des maladies pulmonaires chroniques représentent deux challenges importants pour la prise en charge des patients et l’évaluation des traitements. La surveillance repose principalement sur les données fonctionnelles respiratoires mais l’évaluation morphologique reste un point essentiel pour le diagnostic et l’évaluation de sévérité. Dans la première partie de cette thèse, nous proposons différents modèles pour quantifier la sévérité de pathologies bronchiques chroniques au scanner. Une approche simple par seuillage adaptatif et une méthode plus sophistiquée de radiomique sont évaluées Dans la seconde partie, nous évaluons une méthode d’apprentissage profond pour contourer automatiquement l’atteinte fibrosante de la sclérodermie en scanner. Nous combinons le recalage élastique vers différents atlas morphologiques thoraciques et l’apprentissage profond pour développer un modèle dont les performances sont équivalentes à celles des radiologues. Dans la dernière partie, nous démontrons que l’étude de la déformation pulmonaire en IRM entre inspiration et expiration peut être utilisée pour repérer les régions pulmonaires en transformation fibreuse, moins déformables au cours de la respiration, et qu’en scanner, l’évaluation de la déformation entre des examens successifs de suivi peut diagnostiquer l’aggravation fibreuse chez les patients sclérodermiques. / Disease staging and monitoring of chronic lung diseases are two major challenges for patient care and evaluation of new therapies. Monitoring mainly relies on pulmonary function testing but morphological assessment is a key point for diagnosis and staging In the first part, we propose different models to score bronchial disease severity on computed tomography (CT) scan. A simple thresholding approach using adapted thresholds and a more sophisticated machine learning approach with radiomics are evaluated In the second part, we evaluate deep learning methods to segment lung fibrosis on chest CT scans in patients with systemic sclerosis. We combine elastic registration to atlases of different thoracic morphology and deep learning to produce a model performing as well as radiologists In the last part of the thesis, we demonstrate that lung deformation assessment between inspiratory and expiratory magnetic resonance images can be used to depict fibrotic lung areas, which show less deformation during respiration and that CT assessment of lung deformation on serial CT scans can be used to diagnose lung fibrosis worsening

Research conducted in the aviation industry includes two major areas, increased safety and a reduction of the environmental footprint. This thesis investigates the possibilities of increased situational awareness with computer vision in avionics systems. Image fusion methods are evaluated with appropriate pre-processing of three image sensors, one in the visual spectrum and two in the infra-red spectrum. The sensor setup is chosen to cope with the different weather and operational conditions of an aircraft, with a focus on the final approach and landing phases. Extensive image quality assessment metrics derived from a systematic review is applied to provide a precise evaluation of the image quality of the fusion methods. A total of four image fusion methods are evaluated, where two are convolutional network-based, using the networks for feature extraction in the detailed layers. Other approaches with visual saliency maps and sparse representation are also evaluated. With methods implemented in MATLAB, results show that a conventional method implementing a rolling guidance filter for layer separation and visual saliency map provides the best results. The results are further confirmed with a subjective ranking test, where the image quality of the fusion methods is evaluated further.