Supervised Classification in Google Earth Engine | Sentinel-2 Burned Area Mapping with Random Forest

In this complete tutorial, I show you the full workflow for burned area mapping using Sentinel-2, Google Earth Engine (GEE), and QGIS.
You will learn how to perform supervised classification in GEE, export the classified results as a shapefile, and then refine and clean the polygons inside QGIS.

Code: https://github.com/gis-rs/gee/blob/77c265f1fff5e4658c872c2b2f95649360d9da9b/Classify%20Burned%20Area%20using%20Random%20Forest%20in%20GEE.js

🔥 What You Will Learn in This Video:

🛰️ Google Earth Engine (GEE) Part
✔ Load and preprocess Sentinel-2 imagery
✔ Create training samples
✔ Apply Random Forest supervised classification
✔ Extract burned areas
✔ Convert raster to vectors (reduceToVectors)
✔ Export burned area polygons as a Shapefile

🗺️ QGIS Post-Processing Part
✔ Import the exported shapefile into QGIS
✔ Calculate polygon area (m² / hectares)
✔ Remove small isolated polygons (noise cleanup)
✔ Smooth polygon boundaries for better visualization
✔ Prepare final burned area map for reporting or analysis

This tutorial is perfect for anyone working with wildfire assessment, land cover mapping, machine learning in remote sensing, and GEE-to-QGIS workflows.
By the end of this video, you will have clean, accurate burned area polygons ready for use in GIS projects, research, or professional mapping.

#googleearthengine #machinelearning #randomforest #qgis #supervisedlearning #remotesensing

Видео Supervised Classification in Google Earth Engine | Sentinel-2 Burned Area Mapping with Random Forest канала GIS & RS Made Easy