2023

Training Machine Learning Models for Cloud Optical Thickness Estimation using Synthetic Data

Olof Mogren, Aleksis Pirinen, Thomas Ohlson Timoudas, Anders Karlsson

Summary

Training Machine Learning Models for Cloud Optical Thickness Estimation using Synthetic Data. 200,000 data points simulating how atmosphere distorts light. Models trained on synthetic data show promising results on real satellite imagery. Partnership with Swedish Forestry Agency (SFA).

Themes

machine-learningearth-observation

Keywords

cloud optical thickness, synthetic data, satellite imagery, atmospheric modeling, remote sensing

Poster

Click image to open full size