Fusion of Sentinel-1A and Sentinel-2A data for land cover mapping: a case study in the lower Magdalena region, Colombia
Additional Document Info
Land cover–land use (LCLU) classification tasks can take advantage of the fusion of radar andoptical remote sensing data, leading generally to increase mapping accuracy. Here wepropose a methodological approach to fuse information from the new European SpaceAgency Sentinel-1 and Sentinel-2 imagery for accurate land cover mapping of a portion ofthe Lower Magdalena region, Colombia. Data pre-processing was carried out using theEuropean Space Agency’s Sentinel Application Platform and the SEN2COR toolboxes. LCLUclassification was performed following an object-based and spectral classification approach,exploiting also vegetation indices. A comparison of classification performance using threecommonly used classification algorithms was performed. The radar and visible-near infraredintegrated dataset classified with a Support Vector Machine algorithm produce the mostaccurate LCLU map, showing an overall classification accuracy of 88.75%, and a Kappacoefficient of 0.86. The proposed mapping approach has the main advantages of combiningthe all-weather capability of the radar sensor, spectrally rich information in the visible-nearinfrared spectrum, with the short revisit period of both satellites. The mapping resultsrepresent an important step toward future tasks of aboveground biomass and carbonestimation in the region.