Análisis multitemporal de la especie vegetal invasora Retamo espinoso (Ulex europaeus) en el embalse la Regadera, zona rural de la localidad de Usme, a partir de imágenes satelitales Sentinel 2 y Landsat 8 mediante el uso de algoritmos de clasificación.

Abstract

In this study three supervised methods are used for the identification of the plant species Spiny Loop (Ulex europaeus), the first method is the Artificial Neural Networks, which allows the parameters to be modified according to the working environment and not in dependence on the statistical distribution of the data, The second method is the vector support machines which minimize the probability of misclassifying a data point of a fixed but unknown probability distribution, and finally, the decision trees, which allow the hierarchical classification of coverages based on knowledge of the spectral properties of each class and the relationships between them. Using Landsat 8 and Sentinel 2 satellite images corresponding to years 2014, 2017 and 2018, the classifications of the species are evaluated and changes are detected for the years mentioned, with the purpose of determining the growth it has had and making a prediction of future growth with the objective of serving as a basis for establishing plans or strategies to control this plant species.