Modelo para clasificación de imágenes multiespectrales a partir de los Complejos de Células Abstractas (CCA)

Abstract

The study and analysis of the physical characteristics of the terrestrial surface realized through the raster images is based on images that are product of the spectral response of each of the components sensed in the scene and that under a statistical procedure, probabilistic or of graphs performs a classification process in which the spectral response receives a numerical value that allows its representation in digital format (representation of the continuous space using finite elements). Current image classification processes are developed based on a representation of the space in which the characteristics described by the topology (adjacency, connectivity, and boundary between objects) are not fulfilled. The present work seeks to develop a model for the image classification process, which is based on a representation of the space that meets the requirements to be considered a discrete topological space from the set of locally finite elements consistent, which meets the axioms Of a topological space with the separation property T0.