CoordenaÃÃo de AperfeiÃoamento de NÃvel Superior / Este trabalho propÃe melhorias em uma tÃcnica de amostragem adaptativa multidimensional para renderizaÃÃo. RenderizaÃÃo à o processo de sÃntese de imagens por meio de algoritmos que simulam a iluminaÃÃo em cenÃrios virtuais. As tÃcnicas mais gerais de renderizaÃÃo fotorrealÃstica â aquelas que procuram obter imagens que se assemelham a fotografias â utilizam mÃtodos de integraÃÃo baseados em Monte Carlo para resolver a equaÃÃo que descreve a distribuiÃÃo de luz na cena (equaÃÃo de renderizaÃÃo). Por ser um mÃtodo probabilÃstico e utilizar amostras geradas randomicamente, Monte Carlo produz ruÃdo na imagem final â resultado da variÃncia das amostras â e portanto, pode necessitar de uma grande quantidade de amostras para que o ruÃdo diminua a nÃveis aceitÃveis. Com o intuito de se obter imagens de melhor qualidade com uma menor quantidade de amostras, foram pospostas tÃcnicas de amostragem adaptativa que visam concentrar o esforÃo de amostragem em regiÃes mais importantes da cena. Neste trabalho, propÃe-se a modificaÃÃo de uma tÃcnica de amostragem adaptativa multidimensional por meio da adiÃÃo de duas etapas: substituiÃÃo de amostras e integraÃÃo auxiliar. Essas etapas visam dar mais robustez à tÃcnica, possibilitando sua utilizaÃÃo em uma maior variedade de situaÃÃes. AlÃm da adiÃÃo de duas etapas, tambÃm propÃe-se uma tÃcnica de reconstruÃÃo mais eficiente na etapa final. / This work proposes improvements in a multidimensional adaptive sampling technique for rendering. Rendering is the process of synthesizing images by algorithms simulating lighting in virtual scenes. The more general techniques of photorealistic rendering â those seeking images that resemble photographs â use integration methods based on Monte Carlo to solve the equation that describes the distribution of light in the scene (rendering equation). Being a probabilistic method which uses randomly generated samples, Monte Carlo produces noise in the final image â result of samplesâ variance â and therefore may require a large amount of samples to reduce the noise to acceptable levels. To obtain images of better quality with a lower number of samples, adaptive sampling techniques were proposed, concentrating sampling effort in the most important regions. In this work, we propose the addition of two steps to a multidimensional adaptive sampling technique: substitution of samples and auxiliary integration. These steps aim to give more strength to the technique, enabling their use in a wider variety of situations.

The bulk of the research work in wave related areas considers sea waves as stochastic objects leading to wave forecasting techniques based on statistical approaches. Due to the complex dynamics of the sea waves’ behaviour, statistical techniques are probably the only viable approach when forecasting over substantial spatial and temporal intervals. However this view changes when limiting the forecasting time to a few seconds or when the goal is to estimate the quiescent periods that occur due to the beating interaction of the wave components, especially in narrow band seas. This work considers the multi disciplinary research field of deterministic sea wave prediction (DSWP), exploring different aspects of DSWP associated with shallow angle LIDAR systems. The main goal of this project is to study and develop techniques to reduce the prediction error. The first part deals with issues related to shallow angle LIDAR systems data problems, while the remaining part of this work concentrates on the prediction system and propagation models regardless of the source of the data. The two main LIDAR data problems addressed in this work are the non-uniform distribution and the shadow region problems. An empirical approach is used to identify the characteristics of shadow regions associated with different wave conditions and different laser position. A new reconstruction method is developed to address the non-uniformed sampling problem, it is shown that including more information about the geometry and the dynamics of the problem improves the reconstruction error considerably. The frequency domain approach to the wave propagation model is examined. The effect of energy leakage on the prediction error is illustrated. Two approaches are explored to reduce this error. First a modification of the simple dispersive phase shifting filter is tested and shown to improve the prediction. The second approach is to reduce the energy leakage with an iterative Window-Expansion method. Significant improvement of the prediction error is achieved using this method in comparison to the End-Matching method typically used in DSWP systems. The final part in examining the frequency domain approach is to define the prediction region boundaries associated with a given prediction accuracy. The second propagation model approach is the Time/Space domain approach. In this method the convolution of the measured data and the propagation filter impulse response is used in the prediction system. In this part of the research work properties of these impulse responses are identified. These are found to be quite complicated representations. The relation between the impulse response (duration and shift) with prediction time and distance are studied. Quantification of these impulse responses properties are obtained by polynomial approximation and non-symmetric filter analysis. A new method is shown to associate the impulse response properties to the prediction region of both the Fixed Time and Fixed Point mode.