This package implements a simple Python interface to the USGS Landsat 8 tool for computing per-pixel solar and sensor azimuth and zenith angles, from Angle Coefficient Files (https://www.usgs.gov/core-science-systems/nli/landsat/solar-illumination-and-sensor-viewing-angle-coefficient-files)
Since this package depends on C code, it is recommended that you either build the package with conda-build, using the recipe at (https://github.com/DHI-GRAS/py-l8angles-conda) or install with the latest binaries from the DHI-GRAS conda channel:
conda config --add channels DHI-GRAS
conda install py-l8angles
The package exposes a single function calculate_angles, taking the following parameters:
- metadata_file: str
Path to the ANG file - angle_type: str (default: 'BOTH')
What angles to compute, either 'BOTH', 'SOLAR' or 'SATELLITE' - subsample: int (default: 1)
Subsample factor, i.e. a subsample of 2 will halve the resolution - bands: list (default: [1..11])
What bands to compute angles for
The function returns a dictionary, mapping angle type (sun/sat_az/zn, where az and zn is azimuth and zenith) to lists of numpy 2D arrays. Each array in a list correspond to a single band. The lists are ordered with respect to the input bands.
import l8angles
data = l8angles.calculate_angles('./test_ANG.txt', angle_type='SOLAR',
subsample=2, bands=[3,6,7])