Nuevas herramientas para el estudio de la interacción entre especies en el espacio y en el tiempo

Las funciones de segundo orden son cada vez más empleadas en el análisis de procesos ecológicos. En este trabajo presentamos dos funciones de 2º orden desarrolladas recientemente que permiten analizar la interacción espacio-temporal entre dos especies o tipos funcionales de individuos. Estas funciones han sido desarrolladas para el estudio de interacciones entre especies en masas forestales a partir de la actual distribución diamétrica de los árboles. La primera de ellas es la función bivariante para procesos de puntos con marca Krsmm, que permite analizar la correlación espacial de una variable entre los individuos pertenecientes a dos especies en función de la distancia. La segunda es la función de reemplazo , que permite analizar la asociación entre los individuos pertenecientes a dos especies en función de la diferencia entre sus diámetros u otra variable asociada a dichos individuos. Para mostrar el comportamiento de ambas funciones en el análisis de sistemas forestales en los que operan diferentes procesos ecológicos se presentan tres casos de estudio: una masa mixta de Pinus pinea L. y Pinus pinaster Ait. en la Meseta Norte, un bosque de niebla de la Región Tropical Andina y el ecotono entre las masas de Quercus pyrenaica Willd. y Pinus sylvestris L. en el Sistema Central, en los que tanto la función Krsmm como la función r se utilizan para analizar la dinámica forestal a partir de parcelas experimentales con todos los árboles localizados y de parcelas de inventario.

Geometric 3D point cloud compression

The use of 3D data in mobile robotics applications provides valuable information about the robot’s environment but usually the huge amount of 3D information is unmanageable by the robot storage and computing capabilities. A data compression is necessary to store and manage this information but preserving as much information as possible. In this paper, we propose a 3D lossy compression system based on plane extraction which represent the points of each scene plane as a Delaunay triangulation and a set of points/area information. The compression system can be customized to achieve different data compression or accuracy ratios. It also supports a color segmentation stage to preserve original scene color information and provides a realistic scene reconstruction. The design of the method provides a fast scene reconstruction useful for further visualization or processing tasks.

A three-dimensional representation method for noisy point clouds based on growing self-organizing maps accelerated on GPUs

The research described in this thesis was motivated by the need of a robust model capable of representing 3D data obtained with 3D sensors, which are inherently noisy. In addition, time constraints have to be considered as these sensors are capable of providing a 3D data stream in real time. This thesis proposed the use of Self-Organizing Maps (SOMs) as a 3D representation model. In particular, we proposed the use of the Growing Neural Gas (GNG) network, which has been successfully used for clustering, pattern recognition and topology representation of multi-dimensional data. Until now, Self-Organizing Maps have been primarily computed offline and their application in 3D data has mainly focused on free noise models, without considering time constraints. It is proposed a hardware implementation leveraging the computing power of modern GPUs, which takes advantage of a new paradigm coined as General-Purpose Computing on Graphics Processing Units (GPGPU). The proposed methods were applied to different problem and applications in the area of computer vision such as the recognition and localization of objects, visual surveillance or 3D reconstruction.