This thesis represents two approaches for three dimensional face recognition. In the first approach, a generic face model is fitted to human face. Local shape descriptors are located on the nodes of generic model mesh. Discriminative local shape descriptors on the nodes are selected and fed as input into the face recognition system. In the second approach, local shape descriptors which are uniformly distributed across the face are calculated. Among the calculated shape descriptors that are discriminative for recognition process are selected and used for three dimensional face recognition. Both approaches are tested with widely accepted FRGCv2.0 database and experiment protocol. Reported results are better than the state-of-theart systems. Recognition performances for neutral and non-neutral faces are also reported.

Laboratory experiments were performed to study and improve longshore sediment transport rate predictions. Measured total longshore transport in the laboratory was approximately three times greater for plunging breakers than spilling breakers. Three distinct zones of longshore transport were observed across the surf zone: the incipient breaker zone, inner surf zone, and swash zone. Transport at incipient breaking was influenced by breaker type; inner surf zone transport was dominated by wave height, independent of wave period; and swash zone transport was dependent on wave period. Selected predictive formulas to compute total load and distributed load transport were compared to laboratory and field data. Equations by Kamphuis (1991) and Madsen et al. (2003) gave consistent total sediment transport estimates for both laboratory and field data. Additionally, the CERC formula predicted measurements well if calibrated and applied to similar breaker types. Each of the distributed load models had shortcomings. The energetics model of Bodge and Dean (1987) was sensitive to fluctuations in energy dissipation and often predicted transport peaks that were not present in the data. The Watanabe (1992) equation, based on time-averaged bottom stress, predicted no transport at most laboratory locations. The Van Rijn (1993) model was comprehensive and required hydrodynamic, bedform, and sediment data. The model estimated the laboratory cross-shore distribution well, but greatly overestimated field transport. Seven models were developed in this study based on the principle that transported sediment is mobilized by the total shear stress acting on the bottom and transported by the current at that location. Shear stress, including the turbulent component, was calculated from the wave orbital velocity. Models 1 through 3 gave good estimates of the transport distribution, but underpredicted the transport peak near the plunging wave breakpoint. A suspension term was included in Models 4 through 7, which improved estimates near breaking for plunging breakers. Models 4, 5 and 7 also compared well to the field measurements. It was concluded that breaker type is an important variable in determining the amount of transport that occurs at a location. Lastly, inclusion of the turbulent component of the orbital velocity is vital in predictive sediment transport equations.

Localization is an important problem that must be resolved in order for a robot to make an estimation of its location based on observation and odometry updates. Relying on artificial landmarks such as the lines, circles, and goalposts in the robot soccer domain, current robot localization requires prior knowledge and suffers from uncertainty problems due to partial observation, and thus is less generalizable compared to human beings, who refer to their surroundings for complimentary information. To improve the certainty of the localization model, we propose a framework that recognizes orientation by actively using natural landmarks from the off-field surroundings, extracting these visual features from raw images. Our approach involves identifying visual features and natural landmarks, training with localization information to understand the surroundings, and prediction based on matching of features. This approach can increase the precision of robot orientation and improve localization accuracy by eliminating uncertain hypotheses, and in addition, it is also a general approach that can be extended and applied to other localization problems as well. / text

Sea water intake structures submerged in the surf zone are used to provide water for cooling processes in large facilities such as power plants and refineries. Structures submerged in the surf zone are subject to large forces from breaking waves. To study these forces induced from realistic sea state conditions, a physical model of an intake structure submerged in the wave breaking zone was constructed and subjected to a wide spectrum of regular and irregular waves. The model structure was designed in a manner so force measurement could be isolated to separate components of the structure. The data of peak forces on the structure was analyzed for correlations with varying irregular wave properties. Using the results of forcing on the structure from regular wave tests, drag and inertia coefficients for use in the Morison equation were determined for each separate component and configuration of the structure. These force coefficients were plotted against various wave properties to analyze correlations with wave conditions. Finally, the force coefficients for the structure were used with the Morison equation and current data from the experiments to successfully model forcing on the structure during irregular wave tests.

New laboratory data are presented on the influence of long waves on sediment transport in the surf zone. Due to the very significant difficulties in isolating the morphodynamic processes induced by long waves in field conditions, the laboratory study was designed practically to measure the net sediment transport rates, and gradients in sediment transport, arising from the interaction between long waves and short waves in the surf zone. The bathymetric evolution of model sand beaches, with dB50B = 0.2 mm, was observed under monochromatic short waves, long-wave short-wave combinations (free long waves), and bichromatic wave groups (forced long waves). The beach profile change and net cross-shore transport rates, Q(x), were extracted and compared for conditions with and without long waves. The experiments include a range of wave conditions, e.g. high-energy, moderate-energy, low-energy waves, and the beaches evolve to form accretionary, erosive, and intermediate beach states. Hydrodynamic measurements were made to identify the influence of long waves on short waves and to determine the correlation between surf zone bars and standing long waves. A shallow water wave model was modified for this application to surf zone morphodynamics and compared to both hydrodynamics and measured sediment transport. This data clearly demonstrate that free large-amplitude long waves influence surf zone morphodynamics not only under accretive conditions, by promoting onshore sediment transport, but also under erosive conditions, by decreasing offshore transport. For the dominant berm-bar feature, the strong surf beat induces offshore transport in the inner surf zone and onshore transport around the outer surf zone and throughout the shoaling zone. In contrast, forced (bound) long waves and wave groups correlated with bichromatic short wave groups play a pronounced role under erosive conditions, increasing offshore sediment transport across the whole beach profile. For accretionary conditions, only a very narrowbanded wave group promotes onshore sediment transport across the whole beach profile, while broader banded wave groups again promote offshore transport. The modified numerical model of Li et al. (2002) provides good predictions of the standing long wave pattern for the long-wave short-wave combinations, but generally poor agreement for the bichromatic wave groups. Similarly, this model performs poorly in terms of predicting the net sediment transport for all waves, even after optimising the sediment transport coefficients. This is because the model cannot predict the correct hydrodynamics around the breakpoint position and does not correctly represent net sediment transport mechanics. Overall, the model does not correctly predict the trends in beach profile evolution induced by the long waves and wave groups. Further, there is little evidence that the long wave nodal structure plays a dominant role. The influence of the free long waves and wave groups is consistent with the concept of the Gourlay parameter, H/wBsBT, as a dominant parameter controlling net erosion or accretion. Free long waves tend to reduce H/wBsBT, promoting accretion, while wave groups tend to increase H/wBsBT, promoting erosion.

The nearshore current field, defined here as the residual horizontal flow after averaging over the incident wave period, exhibits variability at a range of time and space scales. Some of the variable currents are low frequency gravity wave motions. However, variable, rotational (in the sense of possessing vertical vorticity) flow can also exist as part of the overall nearshore current field. A field and numerical modelling investigation of these variable rotational currents has been carried out. Drifters, which were developed for surfzone use, enabled measurement of the nearshore current structure; the design and testing of these new instruments is described. Two sets of field measurements, using the new drifters and Eulerian instruments were carried out for conditions with swell perpendicular to a plane beach and in strong longshore currents. In the perpendicular swell conditions, an interesting and well-defined feature of the measured trajectories was the development of transient rip currents. Discrete vortices were also observed. In the longshore current case, trajectories with the longshore current displacement removed had complex meandering paths. Lagrangian data were used to make estimates of length scales and dispersion, both of which provide strong evidence that the current field cannot be due to low frequency gravity waves alone. Under the assumption of equipartition of kinetic and potential energy for low frequency gravity waves, Eulerian measurements of velocities and pressure show significant energy due to non-divergent, rotational flow in both the perpendicular swell and longshore current case. A numerical model that can simulate horizontal flow with a directionally spread, random wave field incident on a plane beach was implemented. The model developed transient rip currents that are qualitatively very similar to those seen in the drifter trajectories from the field. The number and intensity of rip currents in the model depended on the beach slope and incident wave spectra. The energy content and cross-shore flux (and hence transport of material) of the rotational current flow component in the simulated flow fields is comparable to that due to low frequency gravity waves. The modelling also provided some evidence that there may be universal characteristics of the rotational currents. The field results and modelling show that variable rotational currents are ubiquitous in the field even when longshore currents and hence shear waves are not present. The term “infragravity turbulence” is suggested to describe the general class of nearshore hydrodynamics not directly associated with shear waves, which is largely disorganised, but contains well defined features such as transient rips currents and large scale horizontal vortices. The results have important implications in the understanding of the transport of material, including sediment, biological material, pollution, and sometimes bathers, in the nearshore zone.

É alarmante a velocidade à qual o mundo à nossa volta está mudando, e é impossível negar que uma das correntes que impulsiona esta mudança no consumo são as preocupações ambientais e a antecipação do estrago que se fez no mundo. Os consumidores estão todos os dias, e todos os minutos, cada vez mais conscientes que se nós continuarmos a consumir ao ritmo de hoje o futuro do nosso mundo não será feliz. Contudo, não existe certeza de como os consumidores agem sobre esta realidade, e até que ponto cada um se importa com este facto. Somos todos diferentes, independentemente da nossa cultura, da nossa criação ou da nossa idade, cada pessoa está marcada pela sua personalidade distinta formada pelos seus valores que por sua vez definem as suas decisões diárias. Conseguimos perceber valores diferentes nas marcas que nos rodeiam, elas são também uma ferramenta de comunicação para o mundo, que reflete quem cada pessoa é, e quem quer ser. Esta pesquisa procura entender se as compras verdes são influenciadas pelos seus próprios valores e os valores que percebem da marca, e até que ponto são moderadas pelos valores verdes de cada individuo. Esta pesquisa foi realizada dentro de um contexto de uma marca de surfe, Quiksilver, pois estes tipos de marcas já mostraram ser ecológicas e têm uma reputação de inspirar consciência ambiental nos consumidores. Este modelo foi testado usando um estudo empírico executado online disponível a habitantes dos Estados Unidos, seguido de uma análise fatorial e modelação de equações estruturais. Foi possível mostrar um efeito moderador dos valores verdes em intenção de compra nas dimensões de abertura à mudança e Auto transcendência. Foi possível mostrar, também, que compras verdes podem ocorrer por distinção social, como foi o caso da dimensão de auto valorização. Este estudo pretende contribuir para a teoria de valores pessoais e comportamento do consumidor, dentro do âmbito de assuntos ambientais, e espera permitir uma melhor compreensão do futuro de comportamento do consumidor. / It is alarming the pace at which the world around us is changing, and it is undeniable that the currents that bring change towards consumption are the environmental issues and worries that we have inflicted upon the world. Consumers are every day more conscious that if we consume at the rate and way of nowadays our world will not have a bright future. However, doubt remains to what extent we as consumers act upon this, and to what point we care. Everyone is different, independent of culture, upbringing or age, everyone has a distinct personality moulded by values that define everyday decisions. We can also perceive different values from brands around us, they are a tool of our communication to the world and we reflect who we are, and who we want to be, on the brands we carry. This research aimed to understand if green purchases were influenced by each individual’s green values, and to what extent their own personal values and the values they perceived from a brand were relevant. This was done in the context of a surf brand, Quiksilver, because these brands have shown to be ecological and have a reputation of inspiring ecological consciousness on consumers. The model was tested by holding an empirical study with an online survey available to residents in the United States of America, and then applying factorial analysis and structural equation modelling. It was possible to show the moderation effect that green values have on purchase intention, in some value dimensions namely openness to change and selftranscendence. It was also possible to show that green purchases may occur for social distinction, as was the case with self-enhancement. This study aimed to contribute to values theory and consumer behaviour, within environmental issues, and allow for a better understanding of the future of consumer behaviour.

Dissertação (mestrado) - Universidade Federal de Santa Catarina, Tecnológico. Programa de Pós-Graduação em Engenharia Ambiental. / Made available in DSpace on 2012-10-21T23:03:17Z (GMT). No. of bitstreams: 1 204191.pdf: 1146735 bytes, checksum: fe40c2d0ec4ca8df3815b93587b7cc5a (MD5) / A indústria do surfe no Brasil e no mundo vem, há mais de 30 anos, gerando resíduos tóxicos e inflamáveis em todos seus processos produtivos e pós-consumo, que são depostos em aterros ou #lixões# sem qualquer tipo de tratamento específico. Estes resíduos classificados pela NBR 10.004 como Classe I são considerados perigosos,

Dissertação (mestrado) - Universidade Federal de Santa Catarina, Centro Tecnológico. Programa de Pós-Graduação em Engenharia Ambiental. / Made available in DSpace on 2012-10-20T10:07:46Z (GMT). No. of bitstreams: 1 193794.pdf: 468479 bytes, checksum: ae55f41098ae8e04d0c669c3ada6222f (MD5) / A crise ambiental que afeta o planeta produz miséria e exclusão social além da destruição da biodiversidade. Os ambientes costeiros sofrem, de forma mais acentuada, os efeitos da crise ambiental em virtude de sua elevada densidade populacional. A crise agrava-se a cada dia devido à incapacidade humana de participar de forma qualificada do processo de gestão ambiental na construção da sustentabilidade. Para combater esse quadro insustentável foi elaborada esta pesquisa, com caráter participativo, a qual foi desenvolvida na Bacia Hidrográfica da Lagoa da Conceição, na cidade de Florianópolis, no Estado de Santa Catarina, baseada em experiências práticas desenvolvidas pelo pesquisador em parceria com a Associação dos Surfistas e Amigos da Praia Mole. Foram utilizadas para atingir-se os objetivos propostos, técnicas de abordagem qualitativa. O objetivo deste estudo é apresentar uma proposta de participação dos surfistas na construção do processo da Sustentabilidade, tendo como base a Gestão ambiental nas zonas costeiras para atuação nos comitês de bacias hidrográficas. A finalidade do estudo é propiciar a organização social dos surfistas através do estímulo no surgimento de associações, usando o surfe como um atrator para a mobilização comunitária, formando capital social para participar do processo de Gestão participativa dos comitês de bacias hidrográficas e proteger os ecossistemas da zona costeira. A síntese dos dados e sua análise possibilitaram o surgimento de propostas de criação e fortalecimento das associações de surfe para participação no processo de Gestão participativa dos comitês de bacias hidrográficas da zona costeira, de Gestão ambiental e geração de consciência para Sustentabilidade nos campeonatos de surfe.

Submitted by Livia Mello (liviacmello@yahoo.com.br) on 2016-09-21T12:42:19Z No. of bitstreams: 1 DissCM.pdf: 2254055 bytes, checksum: 6daee0adfcc56d2fa96093e529bc319c (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-21T18:34:15Z (GMT) No. of bitstreams: 1 DissCM.pdf: 2254055 bytes, checksum: 6daee0adfcc56d2fa96093e529bc319c (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-21T18:34:20Z (GMT) No. of bitstreams: 1 DissCM.pdf: 2254055 bytes, checksum: 6daee0adfcc56d2fa96093e529bc319c (MD5) / Made available in DSpace on 2016-09-21T18:34:27Z (GMT). No. of bitstreams: 1 DissCM.pdf: 2254055 bytes, checksum: 6daee0adfcc56d2fa96093e529bc319c (MD5) Previous issue date: 2015-12-09 / Não recebi financiamento / Perform fast and accurate recognition of objects is still a challenging task in many applications. In this sense the techniques based on detection and description of image features and in correspondence (matching) has been set up as an important technological evolution in the development of object recognition systems. The most common procedure for object recognition systems is that, given some knowledge about the appearance of some objects, one or more images are examined to determine whether the object is present and the location thereof. Recognition of flat objects, such as images and logos is part of that context today and the approaches proposed for vehicle logos recognition problem. This problem has been one of the domain issues in automatic identification solutions. But realize the recognition of logos vehicles in quick way to images while maintaining the precision of the method employed is still a problem. On the other hand there is also the problem of enabling the application of the method speeded Up Robust Features (SURF) in recognition solutions objects. This method makes it possible to identify points of interest in images quickly and robustly. This paper proposes the development of an approach based on image processing and computer vision techniques to carry out reconnaissance vehicles of logos. The proposed method makes use of the SURF algorithm to perform recognition logos vehicles, combining algorithm with other techniques, such as the Canny method, Conci, Carvalho and Rauber, ROI, KNN and RANSAC to solve the problem, performing recognition the logos of vehicles from 45% to 100% of the images analyzed in a good runtime. The development of this research demonstrate a possible alternative choice for the logos of recognition problem of vehicles through efficient use of a model SURF method. / Executar o reconhecimento rápido e preciso de objetos ainda é uma tarefa desafiadora em muitas aplicações. Nesse sentido as técnicas baseadas em detecção e descrição de características de imagens e em correspondência (matching) tem se configurado como uma importante evolução tecnológica no desenvolvimento de sistemas de reconhecimento de objetos. O procedimento comum da maioria dos sistemas de reconhecimento de objetos é que, dado algum conhecimento sobre a aparência de certos objetos, uma ou mais imagens são examinadas para avaliar se o objeto esta presente e a localização do mesmo. Reconhecimento de objetos planos, como imagens e logomarcas insere-se nesse contexto bem como as abordagens atualmente propostas para o problema de reconhecimento de logomarcas de veículos. Este problema tem sido um dos temas de domínio em soluções de identificação de veículos. Porém realizar o reconhecimento de logomarcas de veículos em imagens de forma rápida mantendo a precisão do método empregado ainda é um problema. Por outro lado há também o problema de viabilizar a aplicação do método Speeded Up Robust Features (SURF) em soluções de reconhecimento de objetos. Este método possibilita identificar pontos de interesse em imagens de forma rápida e robusta. O presente trabalho propõe o desenvolvimento de uma abordagem baseada em técnicas de processamento de imagens e visão computacional para realizar o reconhecimento de logomarcas de veículos. O método proposto faz uso do algoritmo SURF para realizar o reconhecimento de logomarcas de veículos, combinando o algoritmo com outras técnicas, tais como o método de Canny, Conci, Carvalho e Rauber, ROI, KNN e RANSAC para solução do problema, realizando o reconhecimento de logomarcas de veiculos entre 45 % a 100 % das imagens analisadas em um bom tempo de execução. O desenvolvimento deste trabalho de pesquisa possibilitou demonstrar uma opção alternativa para o problema de reconhecimento de logomarcas de veículos através de um modelo de utilização eficiente do método SURF.