This work presents two novel methods of estimating the state of a dynamic system in a Kalman Filtering framework. The first is an application specific method for use with systems performing Visual Odometry in a mostly planar scene. Because a Visual Odometry method inherently provides relative information about the pose of a platform, we use this system as part of the time update in a Kalman Filtering framework, and develop a novel way to propagate the uncertainty of the pose through this time update method. Our initial results show that this method is able to reduce localization error significantly with respect to pure INS time update, limiting drift in our test system to around 30 meters for tens of seconds. The second key contribution of this work is the Manifold EKF, a generalized version of the Extended Kalman Filter which is explicitly designed to estimate manifold-valued states. This filter works for a large number of commonly useful manifolds, and may have applications to other manifolds as well. In our tests, the Manifold EKF demonstrated significant advantages in terms of consistency when compared to other filtering methods. We feel that these promising initial results merit further study of the Manifold EKF, related filters, and their properties.

No description available.

Sleep is very important for health and academic performance. The amount of sleep is variable and depends among other things, on the age of the individual. Today there are many varieties of motion sensors that can be used in healthcare and clinical research for sleep registration, such as Actigraph GT9X Link and Actiwatch 2. The purpose of this study was to compare these two sensors, Actigraph and Actiwatch, for sleep registration by analyzing different sleep variables. The more precise sensor of the two will be used in the Energymetabolic laboratory in the case of survey or research. The method involved participants wearing both sensors for two nights and one day and trying to live as normal as possible. A sleep diary was used to complete the registrations, with related questions about sleep. There were 17 healthy participants between the ages 21 and 31. The results showed that the registrations from both sensors have varying spread and deviations. Overall the results were similar between the sensors and the sleep diary, and it was shown that there is no significant difference between the sensors. However, Actigraph deviated on most occasions from the diary. The Energymetabolic laboratory have access to many Actigraph GT9X Link sensors. This means that an evaluation has to be made to decide whether the cost of replacing the Actigraph GT9X Link sensors with Actiwatch 2 is worth it for the small improvement in precision.

<p>This thesis presents a family of automatic elastic registration methods applicable to single and multimodal images of similar or dissimilar dimensions. These registration algorithms employ a generic dynamic linear elastic continuum mechanics model of the tissue deformation which is discretized using the finite element method. The dynamic deformation model provides spatial and temporal correlation between images acquired from different orientations at different times. First, a volumetric registration algorithm is presented which estimates the deformation field by balancing internal deformation forces of the elastic model against external forces derived from an intensity-based similarity measure between images. The registration is achieved by iteratively solving a reduced form of the dynamic deformation equations in response to image-derived nodal forces. A general approach for automatic deformable image registration is also presented in this thesis which deals with different registration problems within a unified framework irrespective of the image modality and dimension. Using the dynamic deformation model, the problem of deformable image registration is approached as a classical state estimation problem with various image similarity measures providing an observation model. With this formulation, single and multiple-modality, 3D-3D and 3D-2D image registration problems can all be treated within the same framework.The registration is achieved through a Kalman-like filtering process which incorporates information from the deformation model and an observation error computed from an intensity-based similarity measure. Correlation ratio, normalized correlation coefficient, mutual information, modality independent neighborhood descriptor and sum of squared differences between images are similarity/distance measures employed for single and multiple modality image registration in this thesis</p> / Doctor of Philosophy (PhD)

Les imageurs des hôpitaux produisent de plus en plus d'images 3D et il y a un nombre croissant d'études de cohortes. Afin d'ouvrir la voie à des méthodes utilisant de larges bases de données, il est nécessaire de développer des approches permettant de rendre ces bases cohérentes en recalant les images. Les principales méthodes actuelles de recalage de groupes utilisent des données denses (voxels) et sélectionnent une référence pour mettre en correspondance l'ensemble des images. Nous proposons une approche de recalage par groupes, sans image de référence, en utilisant seulement des points d'intérêt (Surf3D), applicable à des bases de plusieurs centaines d'images médicales. Nous formulons un problème global fondé sur l'appariement de points d'intérêt. La variabilité inter-individu étant grande, le taux de faux positifs (paires aberrantes) peut être très important (70\%). Une attention particulière est portée sur l'élimination des appariements erronés. Une première contribution effectue le recalage rigide de groupes d'images. Nous calculons les recalages de toutes les paires d'images. En s'appuyant sur le graphe complet de ces recalages, nous formulons le problème global en utilisant l'opérateur laplacien. Des expérimentations avec 400 images scanner CT 3D hétérogènes illustrent la robustesse de notre méthode et sa vitesse d'exécution. Une seconde contribution calcule le recalage déformable de groupes d'images. Nous utilisons des demi-transformations, paramétrées par des pyramides de B-splines, entre chaque image et un espace commun. Des comparaisons sur un jeu de données de référence montrent que notre approche est compétitive avec la littérature tout en étant beaucoup plus rapide. Ces résultats montrent le potentiel des approches basées sur des points d'intérêt pour la mise en correspondance de grandes bases d'images. Nous illustrons les perspectives de notre approche par deux applications : la segmentation multi-atlas et l'anthropologie. / The ever-increasing amount of medical images stored in hospitals offers a great opportunity for big data analysis. In order to pave the way for huge image groups screening, we need to develop methods able to make images databases consistent by group registering those images. Currently, group registration methods generally use dense, voxel-based, representations for images and often pick a reference to register images. We propose a group registration framework, without reference image, by using only interest points (Surf3D), able to register hundreds of medical images. We formulate a global problem based on interest point matching. The inter-patient variability is high, and the outliers ratio can be large (70\%). We pay a particular attention on inhibiting outliers contribution. Our first contribution is a two-step rigid groupwise registration. In the first step, we compute the pairwise rigid registration of each image pair. In a second step, a complete graph of those registrations allows us to formulate a global problem using the laplacian operator. We show experimental results for groups of up to 400 CT-scanner 3D heterogeneous images highlighting the robustness and speed of our approach. In our second contribution, we compute a non-rigid groupwise registration. Our approach involves half-transforms, parametrized by a b-spline pyramid, between each image and a common space. A reference dataset shows that our algorithm provides competitive results while being much faster than previous methods. Those results show the potential of our interest point based registration method for huge datasets of 3D medical images. We also provide to promising perspectives: multi-atlas based segmentation and anthropology.

Improvements in technology and the Internet have seen a rapid rise in the use of technology in various sectors such as medicine, the courts and banking. The conveyancing sector is also experiencing a similar revolution, with technology touted as able to improve the effectiveness of the land registration process. In some jurisdictions, such as New Zealand and Canada, the paper-based land registration system has been replaced with one in which creation, preparation, and lodgement of land title instruments are managed in a wholly electronic environment. In Australia, proposals for an electronic registration system are under way. The research question addressed by this thesis is what would be the impact on fraud of automating the registration process. This is pertinent because of the adverse impact of fraud on the underlying principles of the Torrens system, particularly security of title. This thesis first charts the importance of security of title, examining how security of title is achieved within the Torrens system and the effects that fraud has on this. Case examples are used to analyse perpetration of fraud under the paper registration system. Analysis of functional electronic registration systems in comparison with the paper-based registration system is then undertaken to reveal what changes might be made to conveyancing practices were an electronic registration system implemented. Whether, and if so, how, these changes might impact upon paper based frauds and whether they might open up new opportunities for fraud in an electronic registration system forms the next step in the analysis. The final step is to use these findings to propose measures that might be used to minimise fraud opportunities in an electronic registration system, so that as far as possible the Torrens system might be kept free from fraud, and the philosophical objectives of the system, as initially envisaged by Sir Robert Torrens, might be met.

Cílem předkládané diplomové práce je pomocí lícování obrazů přesně popsat distorzní pole pro následné odstranění geometrické distorze. Snímky zkreslené geometrickou distorzí pochází z prozařovacího elektronového mikroskopu. První část práce se zabývá zejména teorií spojenou s elektronovou mikroskopií, vznikem geometrické distorze a samotnou obrazovou registrací s důrazem na intenzitní flexibilní metody lícování. Ve druhé části je pak představena vytvořená metoda pro modelování geometrické distorze a lícování obrazů postižených slabou geometrickou distorzí. Vyvinutá metoda je následně otestována na testovacích i reálných datech a srovnána s existujícími popsanými metodami pro obrazovou registraci (například open-source softwarem Elastix).

The right to vote is one of the fundamental rights of citizens in democratic countries. In the United States, however, this action is preceded by the obligation to register. Election registrations are still seen as a limiting element of electoral law. The aim of this work is to find out how the registration was influenced in the past and what influenced the number of registered citizens and voter participation in the presidential elections in 2016. The first part deals with federal electoral registration reforms and looks at the current situation. The second part consists of a research in which the relationship determination is utilized by the Pearson correlation coefficient. The work concludes that the ratio of registered age to eligible voters has not increased, although the demographic curve continues to rise. In the case of factors that may affect the number of registered, the direct relationship has not been established. On the contrary, a direct correlation has been shown in the case of the same factors with the turnout in the presidential election.

No description available.

No description available.