A COMBINATION of NORMALIZED DIFFERENCE VEGETATION INDEX and MULTI-TEMPORAL RADAR VEGETATION INDEX of SENTINEL DATA for LANDUSE/LANDCOVER CLASSIFICATION

Landuse/Landcover (LULC) plays an essential role in managing and monitoring natural resources and the environment in many countries. Optical satellite imagery provides a primary data source for LULC mapping. Unfortunately, the application of the optical image is limited due to the effects of clouds, especially in the tropical area in Vietnam. Microwave imagery can observe during the day and night with nearly all-weather capability. Currently, Sentinel-1 satellite with microwave sensor, operated by the European Space Agency, supplies SAR data with global coverage free of charge. This article presents the combination of the Normalized Difference Vegetation Index (NDVI) of Sentinel-2 data and the multi-temporal Radar Vegetation Index (RVI) of Sentinel-1 data for LULC mapping. The Support Vector Machine (SVM) method was used and the classification accuracy achieved 86. The study area is in Hanoi, Vietnam, with many landcover features and various plants such as fruit trees, paddy rice. As a result, combining the NDVI index and the multi-temporal RVI index can classify the species of trees and help to improve the applicability of Sentinel satellite data in landcover change monitoring. © ACRS 2021.All right reserved.