Supervised classification of LULC using Sentinel-2A and Random Forest

This post is about some notes and steps while doing research: LULC supervised classification using Sentinal-2A dataset,and Random Forest machine learning method.

Pre-process

Before starting process of supervised classification, we need to prepare datesets in SNAP. The dataset we use includs:

• B2,B3,B4,B8,SRB11,SRB12,SRB5,SRB6,SRB7,SRB8A,SRB9

• BI2 (The second Brightness Index)

• NDVI

• REP (Red-Edge Position Index)

• MNDWI

After band select we got file *

Random Forest

Random forest classifier is an ensemble tree-based learning algorithm. The random forest classifier is a set of decision trees from a randomly selected subset of the training set. It aggregates the votes from different decision trees to decide the final class of the test object2.

Process

Classification class

5 classes

• Forest

• Farmland

• Water

• Construction

• Bareland

Create training samples and validation samples

1. Create Vector Data Container