Traditional appearance based face recognition (FR) systems use gray scale images, however recently attention has been drawn to the use of color images. Color inputs have a larger dimensionality, which increases the computational cost, and makes the small sample size (SSS) problem in supervised FR systems more challenging. It is therefore important to determine the scenarios in which usage of color information helps the FR system. In this thesis, it was found that inclusion of chromatic information in FR systems is shown to be particularly advantageous in poor illumination conditions. In supervised systems, a color input of optimal dimensionality would improve the FR performance under SSS conditions. A fusion of decisions from individual spectral planes also helps in the SSS scenario. Finally, chromatic information is integrated into a supervised ensemble learner to address pose and illumination variations. This framework significantly boosts FR performance under a range of learning scenarios.

Background: A pilot study can be an important step in the assessment of an intervention by providing information to design the future definitive trial. Pilot studies can be used to estimate the recruitment and retention rates and population variance and to provide preliminary evidence of efficacy potential. However, estimation is poor because pilot studies are small, so sensitivity analyses for the main trial's sample size calculations should be undertaken. Methods: We demonstrate how to carry out easy-to-perform sensitivity analysis for designing trials based on pilot data using an example. Furthermore, we introduce rules of thumb for the size of the pilot study so that the overall sample size, for both pilot and main trials, is minimized. Results: The example illustrates how sample size estimates for the main trial can alter dramatically by plausibly varying assumptions. Required sample size for 90% power varied from 392 to 692 depending on assumptions. Some scenarios were not feasible based on the pilot study recruitment and retention rates. Conclusion: Pilot studies can be used to help design the main trial, but caution should be exercised. We recommend the use of sensitivity analyses to assess the robustness of the design assumptions for a main trial.

Master of Science / Department of Statistics / Christopher Vahl / Among the most used statistical concepts and techniques, seen even in the most cursory of introductory courses, are the confidence interval, binomial distribution, and sample size estimation. This paper investigates a particular case of generating a confidence interval from a binomial experiment in the case where zero successes are expected. Several current methods of generating a binomial proportion confidence interval are examined by means of large-scale simulations and compared in order to determine an ad-hoc method for generating a confidence interval with coverage as close as possible to nominal while minimizing width. This is then used to construct a formula which allows for the estimation of a sample size necessary to obtain a sufficiently narrow confidence interval (with some predetermined probability of success) using the ad-hoc method given a prior estimate of the probability of success for a single trial. With this formula, binomial experiments could potentially be planned more efficiently, allowing researchers to plan only for the amount of precision they deem necessary, rather than trying to work with methods of producing confidence intervals that result in inefficient or, at worst, meaningless bounds.

The mean absolute deviation is a simple and informative measure of variability. Approximate confidence intervals for mean absolute deviations in one-group and two-group designs are derived and are shown to have excellent small-sample properties under moderate nonnormality. Sample size planning formulas are derived.

Several large-sample confidence intervals for the ratio of independent binomial proportions are compared in terms of exact coverage probability and width. A non-iterative approximate Bayesian interval is derived and its frequency properties are superior to all of the non-iterative confidence intervals considered. The approximate Bayesian interval, which is very easy to compute, has performance characteristics that are very similar to the computationally intensive score method. Two sample size determination formulas are presented, one for desired absolute precision and the other for desired relative precision.

Généralement, dans les situations d’hypothèses multiples on cherche à rejeter toutes les hypothèses ou bien une seule d’entre d’elles. Depuis quelques temps on voit apparaître le besoin de répondre à la question : « Peut-on rejeter au moins r hypothèses ? ». Toutefois, les outils statisques pour répondre à cette question sont rares dans la littérature. Nous avons donc entrepris de développer les formules générales de puissance pour les procédures les plus utilisées, soit celles de Bonferroni, de Hochberg et de Holm. Nous avons développé un package R pour le calcul de la taille échantilonnalle pour les tests à hypothèses multiples (multiple endpoints), où l’on désire qu’au moins r des m hypothèses soient significatives. Nous nous limitons au cas où toutes les variables sont continues et nous présentons quatre situations différentes qui dépendent de la structure de la matrice de variance-covariance des données. / Generally, in multiple endpoints situations we want to reject all hypotheses or at least only one of them. For some time now, we see emerge the need to answer the question : "Can we reject at least r hypotheses ?" However, the statistical tools to answer this new problem are rare in the litterature. We decide to develop general power formulas for the principals procedures : Bonferroni’s, Hochberg’s and Holm’s procedures. We also develop an R package for the sample size calculation for multiple endpoints, when we want to reject at least r hypotheses. We limit ourselves in the case where all the variables are continuous and we present four different situations depending on the structure of the data’s variance-covariance matrix.

O Estado de São Paulo é o principal produtor de uvas de mesa do país, sendo a Niágara Rosada (Vitis Labrusca) a cultivar predominante. Para se oferecer produtos de qualidade ao mercado, os produtores necessitam determinar, periodicamente, o estado de maturação das uvas, sendo o teor de sólidos solúveis (tss), a principal variável avaliada. Para essa determinação, utilizase a amostragem dos frutos na área cultivada. O presente trabalho discute, assim, os métodos de amostragem estratificado e aleatório, e o tamanho adequado da amostra de baga individual, para avaliar o estado de maturação da uva Niágara Rosada, com base no teor de sólidos solúveis. O tamanho adequado da amostra de baga individual foi encontrado para os dois métodos de amostragem, separadamente, utilizando-se os métodos Máxima Curvatura, Máxima Curvatura Modificado e Curva da Variancia. Os métodos de amostragem foram comparados utilizando-se uma análise univariada para dados com medidas repetidas, através dos procedimentos GLM e MIXED do SAS. Foram utilizados dois procedimentos, para que se produzisse resultados confiáveis. Os tamanhos mínimos de amostra de baga individual requeridos, para os métodos estratificado e aleatório foram aproximadamente 30 e 27 bagas por área, respectivamente. Os métodos de amostragem estudados apresentaram diferença significativa, e o método aleatório apresentou grande variação máxima e mínima por planta, devendo assim, ser evitado para esse tipo de estudo. / Sao Paulo state is the main table grape producer in Brazil, being the Niágara Rosada (Vitis Labrusca) the predominant cultivar. To offer quality products to the market, the producers need to determine, periodically, the grapes maturation state, being the content of soluble solids the main variable measured. To determine this content, a sample of fruits in an area is collected. This work approaches the random and the stratified sampling methods and the appropriate sample size of individual berry to evaluate the maturation state of the Niágara Rosada based on the content of soluble solids. The appropriate sample size for individual berry was obtained for two sampling methods, separately, using the Maximum Curvature, Modified Maximum Curvature and Variance Curve methods. The sampling methods were compared using a univariate analysis for repeated measures data using the SAS GLM and MIXED procedures. Two different procedures were used to attain reliable results. The minimum berry sample size required for stratified and random methods were approximately 30 and 27 berries by area, respectively. The sampling methods investigated present significantly different results, and the random method presented high maximum and minimum variation by plant and should be avoided for this kind of study.

This study sought to determine the effect upon Tukey's Honestly Significant Difference (HSD) statistic of concurrently violating the assumptions of homogeneity of variance and equal sample sizes. Two forms for the unequal sample size problem were investigated. Kramer's form and the harmonic mean approach were the two unequal sample size procedures studied. The study employed a Monte Carlo simulation procedure which varied sample sizes with a heterogeneity of variance condition. Four thousand experiments were generated. Findings of this study were based upon the empirically obtained significance levels. Five conclusions were reached in this study. The first conclusion was that for the conditions of this study the Kramer form of the HSD statistic is not robust at the .05 or .01 nominal level of significance. A second conclusion was that the harmonic mean form of the HSD statistic is not robust at the .05 and .01 nominal level of significance. A general conclusion reached from all the findings formed the third conclusion. It was that the Kramer form of the HSD test is the preferred procedure under combined assumption violations of variance heterogeneity and unequal sample sizes. Two additional conclusions are based on related findings. The fourth conclusion was that for the combined assumption violations in this study, the actual significance levels (probability levels) were less-than the nominal significance levels when the magnitude of the unequal variances were positively related to the magnitude of the unequal sample sizes. The fifth and last conclusion was that for the concurrent assumption violation of variance heterogeneity and unequal sample sizes, the actual significance levels significantly exceed the nominal significance levels when the magnitude of the unequal variances are negatively related to the magnitude of the unequal sample sizes.

A curva de retenção da água no solo é um dos principais instrumentos para avaliar a qualidade física dos solos e possibilitar seu manejo adequado. Por meio da Teoria da Capilaridade vários equipamentos foram desenvolvidos para determinar a intensidade com que a água está retida ao solo, porém, pouco se tem dado atenção para verificar se os pressupostos para o real funcionamento da teoria estão sendo atendidos. Um aspecto refere-se ao tamanho da amostra utilizada para determinar a curva de retenção, de modo que haja continuidade dos feixes capilares na amostra e placa porosa. Desta forma, este trabalho propõe avaliar diferentes tamanhos de amostra indeformada para a determinação da curva de retenção. Para isso, coletaram-se amostras em anéis volumétricos cilíndricos de três tamanhos (altura) diferentes (T1 - 0,075 m; T2 - 0,05 m; T3 - 0,025 m;) e mesmo diâmetro interno (0,07 m), dos horizontes diagnósticos de um Latossolo e um Nitossolo em áreas experimentais da Escola Superior de Agricultura Luiz de Queiroz (Esalq/USP), Piracicaba - SP. Realizou-se a caracterização física destes solos, por meio da análise granulométrica, densidade do solo, densidade de partículas, porosidade total e teor de carbono orgânico. As curvas foram determinadas para cada tamanho de amostra, utilizando-se o Funil de Haines, para as tensões 0,5, 1, 4, 6 e 10 kPa, e a Câmara de Pressão de Richards para 33, 100 e 500 kPa. As curvas de retenção foram ajustadas pelo modelo utilizado por van Genuchten. Estimadas as curvas, avaliou-se a distribuição de poros do solo das amostras, determinando-se a curva de frequência acumulada de poros em função do logaritmo do raio e, depois pela diferenciação das equações de ajuste das curvas de retenção, a curva diferencial de frequência acumulada de poros. Os resultados mostram que o Latossolo, por ter textura arenosa no horizonte estudado, não apresentou diferença significativa nas curvas de retenção para os tamanhos das amostras estudadas. Verificou-se pouca modificação na distribuição dos poros deste solo, que possui teor elevado das frações areia fina e muito fina, e desenvolveram papel importante para a retenção de água. O Nitossolo, por sua vez, apresentou diferença significativa da curva obtida pela amostra de menor tamanho (T3), havendo maior retenção de água com a diminuição do tamanho da amostra. Devido a sua textura muito argilosa, o arranjo estrutural deste solo foi diferenciado ao se utilizar as amostras maiores, com provável interrupção e descontinuidade dos feixes capilares. Consequentemente, houve também alteração na distribuição dos poros, com redução dos mesoporos e aumento dos microporos. Desta forma, pode-se concluir que o tamanho da amostra influenciou a curva de retenção da água devido à complexidade estrutural do solo, que provavelmente é diferente nas amostras maiores por causa da continuidade dos feixes capilares, principalmente no Nitossolo. Em outras palavras, quanto menor o tamanho da amostra há menor diferenciação no arranjo de poros, ou seja, maior proximidade da real condição do solo e, assim, uma interpretação da retenção de água \"mais correta\" por meio da Teoria da Capilaridade. / The soil water retention curve is one of main tools to assess the physical quality of the soil and to make possible its adequate management. By means of the Capillary Theory, many instruments have been developed to determine the water retention forces in soil, but, little attention has been given to check whether the assumptions for the application of the theory are being attended. One aspect relates to the sample size used to determine the retention curve, so that there is capillary continuity of the sample and porous plate. Thus, this study aimed to evaluate different sizes of undisturbed cylindrical samples for determination of the retention curve. The samples were collected from diagnostic horizons of Latosol and a Nitosol, in experimental areas of the Escola Superior de Agricultura Luiz de Queiroz (Esalq/USP), Piracicaba - SP. Three volumetric rings with three different heights (T1 - 0,075 m; T2 - 0,05 m; T3 - 0,025 m;), were used the diameter of the rings were the same for the three sizes (0,07 m). The physical characterization soils were made by the granulometric analysis, bulk density, particle density, porosity and organic carbon. The curves were determined for each sample size, using the Haine\'s funnel, for tensions 0.5, 1, 4, 6 and 10 kPa, and Richard\'s pressure chamber for 33, 100 and 500 kPa. The pore size distribution of the soil was evoluated first by determining the cumulative frequency curve of pore radius, and then, by differentiating fitling equation of fluis curve to obtain a differential curve of pore cumulative frequency. The retention curves were fitted by the model used by Van Genuchtem. The results showed that in the Latosol there was no significant difference in retention curves for the sizes of the samples studied. There was little change in pore distribution of the fluis soil with high content of fractions fine and very fine sand, that developes significant role in soil water retention. The Nitosol exhibited significant difference of the curve obtained by the smallest size sample (T3), with greater water retention with the decreasing of the sample size. Because of its very clayey texture, the structural arrangement of this soil was different when larger samples were used, with probable interruption and discontinuity of capillaries. Consequently there was also a change in the pore distribution, with reduction of mesopores and an increase of micropores. Thus, itean be conclude that the sample size influenced the soil water retention curve due to the structural complexity of the soil that probably is different in the bigger sample because of the continuity of the capillary budles fluit was more affected in the Nitosol. In other words, the smaller the sample size, the smaller the difference in the arrangement of pores, that is, more closeness of the actual condition of the soil and so a \"more correct\" interpretation of soil water retention by capillary theory.

Parasite documentation has been an important tool to understand host history and co-evolutionary processes in these associations. Acanthobothrium is a worldwide genus of cestodes, and it is a parasite of sharks, skates and rays, with almost 200 nominal species. Recent studies are presenting new hypotheses on cestodes distribution and host specificity patterns. Due to their large distribution, geographical and in host taxa, Acanthobothrium seems to be a good model to evaluate these hypothesis. The Bay of Marajó is an estuarine area were freshwater stingrays of the family Potamotrygonidae share the same environment with marine dasyatid rays. There is no documentation about the dasyatid fauna of Acanthobothrium for Bay of Marajó, neither their relationships with freshwater lineages or their hosts. In our goal to understand those evolutionary events, our results revealed six new lineages of Acanthobothrium, parasites of Hypanus guttatus and Fontitrygon geijskesi, of which four are formally described. Five of those new lineages are included in a clade exclusive to Northwest Atlantic and Caribbean Sea. Acanthobothrium sp. 10 was recovered as sister of clade formed by marine Acanthobothrium sp. 9 sensu Trevisan and freshwater Acanthobothrium sp. 2 sensu Cardoso Jr. The specificity pattern of Acanthobothrium has been discussed and our results corroborate this discussion when we observed four species of Acanthobothrium sharing two different hosts of different genera. Also, host sample size may be correlated with the specificity and strict specificity patterns of the parasite to their hosts. The absence of Fontitrygon colarensis in our samples and the incongruities in comparison to Hypanus guttatus make us question the taxonomic status of F. Colarensis / A documentação de organismos parasitas tem sido uma ferramenta importante para entender a história de seus hospedeiros e os processos coevolutivos implícitos nessas associações. Acanthobothrium é um gênero de cestóideos mundialmente distribuido, parasita de tubarões, skates e raias, com quase 200 espécies nominais. Estudos recentes vem apresentando novas hipóteses sobre a distribuição e especificidade dos cestóides. Devido à sua larga distribuição geográfica e em taxons hospedeiros, Acanthobothrium parece ser um bom modelo para avaliar estas hipóteses. A Baía de Marajó é uma área estuarina onde arraias de água doce da família Potamotrygonidae compartilham o mesmo ambiente com raias marinhas, especialmente dasiatídeos. Não há documentação sobre a fauna de Dasyatidae para Acanthobothrium a Baía de Marajó, nem sobre as relações destas com linhagens de água doce e seus hospedeiros. Com o objetivo de entender esses eventos evolutivos, nossos resultados mostraram seis novas linhagens de Acanthobothrium, parasitas de Hypanus guttatus e Fontitrygon geijskesi, das quais quatro são descritas. Cinco dessas novas linhagens estão incluídas em um clado exclusivamente do Noroeste Atlântico e do Mar do Caribe. Acanthobothrium sp. 10 foi recuperado como grupo irmão de um clado formado pela linhagem marinha Acanthobothrium sp. 9 sensu Trevisan e pela linhagem de água doce Acanthobothrium sp. 2 sensu Cardoso Jr. O padrão de especificidade de Acanthobothrium tem sido discutido e nossos resultados corroboram esta discussão quando observamos quatro espécies de Acanthobothrium compartilhando duas hospedeiras de diferentes gêneros. Além disso, tamanho amostral de hospedeiros pode estar relacionado com nosso entendimento sobre os padrões de especificidade estrita do parasita aos seus hospedeiros. A ausência de F. Colarensis em nossas amostras e as incongruências em comparação com H. Guttatus nos fazem questionar o status taxonômico de F. Colarensis