This repository contains code for implementing prototype algorithm workflows for the Space-based Imaging Spectroscopy and Thermal pathfindER (SISTER).
This repository is under active development and currently contains code for preprocessing imaging spectroscopy data from airborne and spaceborne sensors for input into higher level algorithms including atmospheric, topographic and BRDF correction algorithms.
We recommend installing the libary and its dependencies in a conda environment.
To create and activate a new environment run:
conda create -n sister python=3.8
source activate sister
Next install gdal:
conda install gdal
To install the library, clone:
git clone https://github.com/EnSpec/sister.git
and install with pip:
pip install ./sister
The following code takes as input a PRISMA L1 radiance image along with ESA Copernicus DEM tiles and exports three ENVI formated files:
- Merged VNIR+SWIR radiance datacube
- Location datacube (longitude, latitude, altitude)
- Observables datacube (sensor, solar geometry......)
import os
from sister.sensors import prisma
l1_zip = '/data/prisma/PRS_L1_STD_OFFL_ 20200621003500_20200621003505_0001.zip'
out_dir = '/data/prisma/rad/'
temp_dir = '/data/temp/'
elev_dir = 'https://copernicus-dem-30m.s3.amazonaws.com/'
prisma.he5_to_envi(l1_zip,
out_dir,
temp_dir,
elev_dir,
shift = './data/prisma/PRISMA_Mali1_wavelength_shift_surface_smooth.npz',
rad_coeff = './data/prisma/PRS_Mali1_radcoeff_surface.npz',
match= True,
proj = True)