In the lectures and reading this week, you've learned about hyperspectral remote sensing and a number of different methods for analyzing hyperspectral data. In this practical, we'll gain some experience working with hyperspectral data, using a few examples written in python.

• Open and view data using xarray
• Perform atmospheric correction using dark object subtraction
• Use spectral angle matching to compare spectral signatures and identify surfaces
• Gain some familiarity with Spectral Python (SPy), a python package for analyzing hyperspectral images.

In the data folder, you should have the following files:

• combine_spectral_library.py - a python script that can be used to add additional spectra to spectral_library.csv
• EO1H0380352003173110PF_MTL_L1T.TXT - the L1T ("level-1 terrain corrected") metadata file for the EO-1 image
• solar_spectra.csv - a csv file containing values of extraterrestrial (top of atmosphere) radiance as a function of wavelength
• spectral_library.csv - a csv file containing example reflectance spectra for different surface types

You'll need to download the hyperspectral data from Blackboard, or from the Google Drive link here - be sure to save it to the data folder so that it can be opened using the code in the notebook.

Once you clone the repository, you can set up the conda environment using the provided environment.yml file.

To get started working through the practical, launch jupyter lab and open the notebook (Hyperspectral Image Analysis.ipynb).