In this work, we present a method for predicting hot spot residues by using a set of structural and evolutionary parameters. Unlike previous studies, we use a set of parameters which do not depend on the structure of the protein in complex, so that the predictor can also be used when the interface region is unknown. Despite the fact that no information concerning proteins in complex is used for prediction, the application of the method to a compiled dataset described in the literature achieved a performance of 60.4%, as measured by F-Measure, corresponding to a recall of 78.1% and a precision of 49.5%. This result is higher than those reported by previous studies using the same data set.

The system is composed of two main modules. The first one, carries out the image prosecution, whose objective is to determine the depth map of a pair of views where each pair of successive views continues a sequence of phases: interest points detection, points correspondence and points reconstruction; in the reconstruction process, is determined the parameters that describe the movement (rotation matrix R and the translation vector T) between the two views. This an sequence of steps is repeated for all the peers of successive views of the set. The second module has as objective to create the 3D model of the object, for it should determine the total map of all the 3D points generated, by each iteration of the previous module, once obtained the map of total depth generates the 3D netting, applying the triangulation method of Delaunay [28]. The results obtained of the reconstruction process are shaped in a VRML virtual environment, to obtain a viewing more realist of the object.

Este trabajo presenta el desarrollo de un sistema de reconstrucción de objeto 3D, a partir de una colección de vistas. El sistema se compone de dos módulos principales. El primero realiza el procesamiento de imagen, cuyo objetivo es determinar el mapa de profundidad en un par de vistas, donde cada par de vistas sucesivas sigue una secuencia de fases: detección de puntos de interés, correspondencia de puntos y reconstrucción de puntos; en el proceso de reconstrucción se determinan los parámetros que describen el movimiento (matriz de rotación R y el vector de traslación T) entre las dos vistas. Esta secuencia de pasos se repite para todos los pares de vista sucesivas del conjunto. El segundo módulo tiene como objetivo crear el modelo 3D del objeto, para lo cual debe determinar el mapa total de todos los puntos 3D generados; en cada iteración del módulo anterior, una vez obtenido el mapa de profundidad total, genera la malla 3D, aplicando el método de triangulación de Delaunay [28]. Los resultados obtenidos del proceso de reconstrucción son modelados en un ambiente virtual VRML para obtener una visualización más realista del objeto.

This work presents a comparison of the calibration techniques proposed by Tuceryan et al. [15], Trucco et al. [13] and Zhang [20], suggesting different ways to determine the parameters, starting from known 3D points and their corresponding points in the 2D image plane. The correspondence is manually extracted, being the obtained parameter accuracies (extrinsic and intrinsic) the most important. The passive patterns used consider the calibration techniques with different homographs, depending on the existing process planes. Zhang [20] proposes an approximation based on patterns located on a 2D planar surface, where the camera or the pattern are displaced without knowledge of the motion. Tuceryan et al. [15] and Trucco et al. [13] base their procedure in a known 3D model and its correspondence on a 2D image plane, the difference being in the consideration of the intrinsic parameters that must be determined or considered in the final camera matrix. The algorithms were implemented using Matlab 6.0, according to the interpretation of the works and optimization techniques extracted from the cited literature.

Este trabajo presenta una comparación de las técnicas de calibración propuestas por Tuceryan et al. [15], Trucco et al. [13] y Zhang [20], los cuales proponen maneras diferentes para determinar los parámetros a partir de puntos 3D conocidos y sus correspondientes en el plano imagen 2D. La correspondencia es extraída en forma manual, centrando la atención en la exactitud de los parámetros de cámara entregados (intrínsecos y extrínsecos). Los patrones pasivos utilizados consideran las técnicas de calibración con diferentes homografías, dependiendo de los planos existentes en el proceso. Zhang [20] propone una aproximación basada en patrones acomodados en una superficie planar 2D, donde la cámara o el patrón se desplaza sin conocer el movimiento. En cambio Tuceryan et al. [15] y Trucco et al. [13] se basan en un modelo conocido 3D y su correspondencia en el plano imagen 2D, diferenciándose en la consideración de los parámetros intrínsecos que deben ser determinados o de influencia en la matriz de cámara final. Los algoritmos fueron implementados en Matlab 6.0, de acuerdo a la interpretación de los trabajos y técnicas de optimización extraídas desde la literatura citada.

Este artigo descreve um método para caracterização de nervuras utilizando a representação polinomial cúbica de Hermite. Os elementos associados a esta representação são usados como descritores das nervuras e o potencial discriminatório é analisado com base na identificação de duas espécies de leguminosas (Lonchocarpus muehlbergianus Hassl. e L. subglaucescens Mart, ex Benth.). Os vetores de geometria de Hermite exercem influência sobre toda a extensão da curva modelada, permitindo a descrição global da nervura por um descritor de baixa dimensionalidade. Os resultados obtidos mostram que o método foi eficaz na separação das espécies analisadas e apresenta-se como uma opção adicional aos métodos para caracterização de nervuras usualmente utilizados em taxonomia vegetal.

This paper describes a method for leaf vein shape characterization using Hermite polynomial cubic representation. The elements associated with this representation are used as secondary vein descriptors and their discriminatory potential are analyzed based on the identification of two legume species (Lonchocarpus muehlbergianus Hassl. and L. subglaucescens Mart, ex Benth.). The elements of Hermite geometry influence a curve along all its extension allowing a global description of the secondary vein course by a descriptor of low dimensionality. The obtained results shown the analyzed species can be discriminated by this method and it can be used in addition to commonly considered elements in the taxonomic process.