Creation of a gala apple fruit image database for Brazil.
No hay miniatura disponible
Fecha
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
Descripción
Information on land use and coverage is necessary to assist in the management process and assertive decision-making. Thus, the present study aimed to evaluate the fusion of Sentinel-1 (S1) and Sentinel-2 (S2) data in the mapping of land use and coverage of the municipality of Lagoinha (SP) using the Random Forest method. Three scenarios were tested for classification: data from (S1), (S2) and fusion of (S2+S1). To evaluate the accuracy of the classification, high-resolution images from Google Earth and S2 software were used. The overall accuracy of the classification from the combination of S2+S1 data was 94%, and the Kappa index was equal to 0.9. For the isolated images of S2 and S1, overall accuracies of 80% and 50% and Kappas index of 0.71 and 0.50 were obtained, respectively. The fusion of S1+S2 data showed high accuracy in mapping;
Palabras clave
Sensor Fusion, Machine Learning, Radar, Remote sensing, Land use