A common approach to evaluating forest conditions is by measuring or estimating tree height, which produces what are known as canopy height maps. Accurate, high-resolution canopy height maps are essential for understanding forest dynamics and cycles, as well as for effective forest management and climate change mitigation. Check out Estimating Canopy Height at Scale to read more.

Galileo is a pre-trained model for remote sensing tasks. It is quite special compared to previously published models, because it was trained in a clever way to be able to process many different types of input data. Check out the paper Galileo: Learning Global & Local Features of Many Remote Sensing Modalities.

The goal of this bachelor thesis is to apply the Galileo model to generate a canopy height map and compare the result with existing canopy height maps.

The tasks for this bachelor thesis consist of:

1. Creating an overview of existing literature and state-of-the-art models for generating canopy height maps,
2. implementing and fine-tuning the Galileo model,
3. generating a canopy height map,
4. evaluating this map and comparing the Galileo approach with other methods.