Abstract
Monitoring greenhouse gas emissions has recently received attention from scientists, policy-makers and agencies. In order to estimate greenhouse gas emissions, an accurate land cover map of the area to be evaluated is essential. This study uses the Random Forest algorithm to classify Landsat 8, Sentinel 1,2 satellite images to build land cover maps according to AFOLU classification schema recommendations, serving greenhouse gas inventories. The remote sensing image classification results in the Northwest and Southwest regions showed an overall accuracy of 80.5%, suitable for building surface coverage maps for greenhouse gas inventory.
PDF (Tiếng Việt)
| Download: 257
Downloads
Download data is not yet available.