Análisis del aporte de las medidas texturales en la identificación del estrés biótico de los frailejones mediante imágenes UAV

Abstract

This work presents a methodology that allows to identify the biotic stress present in frailejones in images obtained by unmanned aerial vehicles (UAV). For this, textural measures of occurrence were used, the co-occurrence matrix (GLCM) and the Fourier transform; and unsupervised and supervised classification algorithms. The results indicate that the combination of spectral information (RGB and NIR) and textural information increase the classification accuracy considerably, that if only the spectral information is considered, the results are more significant when the co-occurrence matrix is used.