This thesis proposes a solution to the problem of estimating the motion of an Unmanned Underwater Vehicle (UUV). Our approach is based on the integration of the incremental measurements which are provided by a vision system. When the vehicle is close to the underwater terrain, it constructs a visual map (so called "mosaic") of the area where the mission takes place while, at the same time, it localizes itself on this map, following the Concurrent Mapping and Localization strategy. The proposed methodology to achieve this goal is based on a feature-based mosaicking algorithm. A down-looking camera is attached to the underwater vehicle. As the vehicle moves, a sequence of images of the sea-floor is acquired by the camera. For every image of the sequence, a set of characteristic features is detected by means of a corner detector. Then, their correspondences are found in the next image of the sequence. Solving the correspondence problem in an accurate and reliable way is a difficult task in computer vision. We consider different alternatives to solve this problem by introducing a detailed analysis of the textural characteristics of the image. This is done in two phases: first comparing different texture operators individually, and next selecting those that best characterize the point/matching pair and using them together to obtain a more robust characterization. Various alternatives are also studied to merge the information provided by the individual texture operators. Finally, the best approach in terms of robustness and efficiency is proposed.After the correspondences have been solved, for every pair of consecutive images we obtain a list of image features in the first image and their matchings in the next frame. Our aim is now to recover the apparent motion of the camera from these features. Although an accurate texture analysis is devoted to the matching pro-cedure, some false matches (known as outliers) could still appear among the right correspon-dences. For this reason, a robust estimation technique is used to estimate the planar transformation (homography) which explains the dominant motion of the image. Next, this homography is used to warp the processed image to the common mosaic frame, constructing a composite image formed by every frame of the sequence. With the aim of estimating the position of the vehicle as the mosaic is being constructed, the 3D motion of the vehicle can be computed from the measurements obtained by a sonar altimeter and the incremental motion computed from the homography.Unfortunately, as the mosaic increases in size, image local alignment errors increase the inaccuracies associated to the position of the vehicle. Occasionally, the trajectory described by the vehicle may cross over itself. In this situation new information is available, and the system can readjust the position estimates. Our proposal consists not only in localizing the vehicle, but also in readjusting the trajectory described by the vehicle when crossover information is obtained. This is achieved by implementing an Augmented State Kalman Filter (ASKF). Kalman filtering appears as an adequate framework to deal with position estimates and their associated covariances.Finally, some experimental results are shown. A laboratory setup has been used to analyze and evaluate the accuracy of the mosaicking system. This setup enables a quantitative measurement of the accumulated errors of the mosaics created in the lab. Then, the results obtained from real sea trials using the URIS underwater vehicle are shown.

El treball desenvolupat en aquesta tesi aprofundeix i aporta solucions innovadores en el camp orientat a tractar el problema de la correspondència en imatges subaquàtiques. En aquests entorns, el que realment complica les tasques de processat és la falta de contorns ben definits per culpa d'imatges esborronades; un fet aquest que es deu fonamentalment a il·luminació deficient o a la manca d'uniformitat dels sistemes d'il·luminació artificials. Els objectius aconseguits en aquesta tesi es poden remarcar en dues grans direccions. Per millorar l'algorisme d'estimació de moviment es va proposar un nou mètode que introdueix paràmetres de textura per rebutjar falses correspondències entre parells d'imatges. Un seguit d'assaigs efectuats en imatges submarines reals han estat portats a terme per seleccionar les estratègies més adients. Amb la finalitat d'aconseguir resultats en temps real, es proposa una innovadora arquitectura VLSI per la implementació d'algunes parts de l'algorisme d'estimació de moviment amb alt cost computacional. / Underwater robotics was the motivation of this work, even though computer vision and parallel VLSI architectures played the most important role. Due to their low cost, high-rate and high-resolution, vision based systems represent a good option to provide information about a vehicle position. The apparent motion of a camera mounted on an underwater vehicle can be estimated by correlating two successive frames of an image sequence. Lack of well-defined contours, as well as non-uniform illumination makes underwater scenes much more difficult to be processed than normal images. Therefore, methods frequently used in standard image processing must be modified and adapted to these particular conditions. A method based on texture characterisation of points to reject outliers from the image correspondence problem is proposed. On the other hand, a parallel implementation was used to speed-up parts of the motion estimation algorithm which have a computationally high load. A new VLSI architecture is proposed with the aim of achieving frame-rate performance.

Thesis (M.S.)--University of Tennessee, Knoxville, 2010. / Title from title page screen (viewed on July 20, 2010). Thesis advisor: Paul Ayers. Vita. Includes bibliographical references.

Orientador:Ricardo Machado Leite de Barros / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Educação Física / Made available in DSpace on 2018-08-21T23:09:57Z (GMT). No. of bitstreams: 1 Silvatti_AmandaPiaia_D.pdf: 2153043 bytes, checksum: bfa39ebf2fe3e82ab53b434c5a4e6d8a (MD5) Previous issue date: 2013 / Resumo: Para análises subaquáticas de movimentos tridimensionais é necessária uma calibração precisa de grandes volumes. Métodos baseados em modelos lineares de câmeras são comumente utilizados na biomecânica e para isso faz-se necessário a construção, o transporte e a medição de estruturas rígidas, o que se torna mais difícil quando volumes maiores estão envolvidos. Recentemente, métodos alternativos baseados em modelos não-lineares de câmeras, foram propostos para resolver este aspecto. Assim, os objetivos deste trabalho foram 1) avaliar a exatidão da reconstrução tridimensional fora e dentro da água utilizando o método proposto por Zhang (2000); 2) avaliar a exatidão dos métodos não-lineares para a calibração de câmeras propostos por Hatze, 1988, Cerveri, et al., 1998 e Zhang, 2000 para aplicações com câmeras submersas e os efeitos da variação da posição no volume de calibração sobre a exatidão de reconstrução tridimensional dos métodos e ressaltando, ainda as vantagens e desvantagens de cada método e 3) testar a aplicabilidade dos métodos não-lineares propostos por Cerveri et al., 1998 e Zhang, 2000 para a reconstrução da trajetória da mão de nadadores em diferentes nados. Para aquisição dos dados foi utilizado um sistema de análise cinemática (DVideo), que foi adaptado para aquisição de imagens submersa. O sistema consiste de computadores ligados em câmeras (Basler) para aquisição online de dados. Caixa-estanques especialmente desenvolvidas para o modelos das câmeras foram utilizadas para proteção e um genlocker trigger foi utilizado para a sincronização das câmeras. Tripés para sustentação e fixação das câmeras foram adaptados com ventosas para fixação no fundo da piscina. Para a correção da distorção foram implementados em Matlab os métodos não-lineares para a calibração de câmeras. Um teste de barra rígida foi usado para avaliação da exatidão e precisão. Os resultados mostraram que os métodos propostos por Cerveri et al., 1998 (1,16mm a 0,96mm) e Zhang, 2000 (0,73mm) são alternativas promissoras para análise tridimensional de movimentos subaquáticos. Ambas as metodologias apresentaram resultados com maior exatidão que os encontrados na literatura. Este fato ocorreu devido a uma melhor modelagem da distorção óptica que foi confirmada pela menor influência no erro relativo à posição do objeto no volume de aquisição. Quanto à flexibilidade e portabilidade em relação ao objeto de calibração, ambos os métodos usam objetos mais fáceis de construir e manipular do que os objetos tradicionalmente utilizados. O sistema para a análise tridimensional do movimento utilizando câmeras submersas mostrou-se ser adequado para aplicações subaquáticas. Uma vez que, os resultados aqui relatados podem ser imediatamente apreciados pelos treinadores, pois foi possível identificar simetria ou assimetria entre os dois lados, a variabilidade intra e inter-sujeitos em termos de padrões de movimento e concordância ou discordância com o modelo teórico. Importante ressaltar que existe a possibilidade de extensão da análise para os diferentes segmentos corporais e o sistema e os métodos para a calibração de câmeras apresentados neste trabalho podem ser utilizado para qualquer esporte ou atividade realizada na água, como por exemplo, hidroginásticas ou práticas de reabilitação / Abstract: In order to perform a three-dimensional motion analysis in underwater conditions is necessary to calibrate accurately large volumes. Methods based on linear camera models are commonly used in biomechanics and this requires to construct, to transport and to measurement rigid structures, which becomes more difficult when larger volumes are involved. Recently, alternative methods based on nonlinear camera models have been proposed to address this aspect. The aims of this study were 1) to evaluate the accuracy of 3D reconstruction out and underwater using the method proposed by Zhang (2000); 2) to evaluate the accuracy of nonlinear camera calibration methods proposed by Hatze, 1988, Cerveri, et al., 1998 and Zhang, 2000 for underwater applications using submerged cameras and the effects of object position in the acquisition volume on the accuracy of 3D reconstruction methods, highlighting the advantages and disadvantages of each method, and 3) to test the applicability of the nonlinear methods proposed by Cerveri et al., 1998 and Zhang, 2000 for the reconstruction of the swimmers hand trajectory in different swims styles. For data acquisition, we used a kinematic analysis system (DVideo), which was adapted for underwater conditions. The system consists of cameras (Basler) connected in computers for online data acquisition. Waterproof housings were specially designed to protect the cameras and a genlocker trigger was used to synchronize the cameras. Tripods to support and to fix the cameras were adapted with suction cups and were used to fix them on the swimming pool floor. In order to perform the distortion correction the nonlinear camera calibration methods were implemented in Matlab software. A rigid bar test was used to assess the accuracy and precision. The results showed that the methods proposed by Cerveri et al. 1 998 (1.16 mm to 0.96 mm) and Zhang, 2000 (0.73 mm) are promising alternatives for 3D underwater motion analysis. Both methodologies presented results with greater accuracy than those found in the literature. This occurred due to an improvement of the distortion modeling and that was confirmed by the smallest influence of the object position on the error values. Related to the flexibility and portability of the calibration object, both methods use objects easier to build and manipulate than the objects traditionally used. The 3D motion analysis system using submerged cameras showed to be suitable for underwater applications. The results reported here can be immediately appreciated by coaches, because it was possible to identify symmetry or asymmetry between the two sides, the variability within and between subjects in terms of movement patterns and agreement or disagreement with the theoretical model. Important to emphasize that there is the possibility of extending the analysis for different body segments and the system and the camera calibration methods presented in this work can be used for any sport or activity performed in water, such as hydro gymnastic or rehabilitation practices / Doutorado / Biodinamica do Movimento Humano / Doutor em Educação Física

Thesis (Ph. D.)--Joint Program in Applied Ocean Science and Engineering (Massachusetts Institute of Technology, Dept. of Ocean Engineering; and the Woods Hole Oceanographic Institution), 2005. / This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Includes bibliographical references (p. 173-187). / This thesis describes a vision-based, large-area, simultaneous localization and mapping (SLAM) algorithm that respects the low-overlap imagery constraints typical of autonomous underwater vehicles (AUVs) while exploiting the inertial sensor information that is routinely available on such platforms. We adopt a systems-level approach exploiting the complementary aspects of inertial sensing and visual perception from a calibrated pose-instrumented platform. This systems-level strategy yields a robust solution to underwater imaging that overcomes many of the unique challenges of a marine environment (e.g., unstructured terrain, low-overlap imagery, moving light source). Our large-area SLAM algorithm recursively incorporates relative-pose constraints using a view-based representation that exploits exact sparsity in the Gaussian canonical form. This sparsity allows for efficient O(n) update complexity in the number of images composing the view-based map by utilizing recent multilevel relaxation techniques. We show that our algorithmic formulation is inherently sparse unlike other feature-based canonical SLAM algorithms, which impose sparseness via pruning approximations. In particular, we investigate the sparsication methodology employed by sparse extended information filters (SEIFs) and offer new insight as to why, and how, its approximation can lead to inconsistencies in the estimated state errors. Lastly, we present a novel algorithm for efficiently extracting consistent marginal covariances useful for data association from the information matrix. / (cont.) In summary, this thesis advances the current state-of-the-art in underwater visual navigation by demonstrating end-to-end automatic processing of the largest visually navigated dataset to date using data collected from a survey of the RMS Titanic (path length over 3 km and 3100 m² of mapped area). This accomplishment embodies the summed contributions of this thesis to several current SLAM research issues including scalability, 6 degree of freedom motion, unstructured environments, and visual perception. / by Ryan M. Eustice. / Ph.D.