The goal of the gms_preprocessing Python library is to provide a fully automatic pre-precessing pipeline for spatial and spectral fusion (i.e., homogenization) of multispectral satellite image data. Currently it offers compatibility to Landsat-5, Landsat-7, Landsat-8, Sentinel-2A and Sentinel-2B.

Feature overview

Level-1 processing:

  • data import and metadata homogenization (compatibility: Landsat-5/7/8, Sentinel-2A/2B)

  • equalization of acquisition- and illumination geometry

  • atmospheric correction (using SICOR)

  • correction of geometric errors (using AROSICS)

Level-2 processing:

  • spatial homogenization

  • spectral homogenization (using SpecHomo)

  • estimation of accuracy layers

=> application oriented analysis dataset