A set of tools and Python libraries to automate various tasks involved in remote sensing.

A tool to sample low and high resolution datasets covering the same area in order to build correlative models mapping one to the other.

Take 10 samples of a Landsat image and a high resolution visible image and write a feature containing the samples:

sample_region.run_sampling(
    registry,
    "ndvi_landsat_13SDU05049011.tif",
    "13SDU050490_201203_0x2000m_CL_1.jp2",
    5)

Take a set of small-scale classified images and build a regression model to estimate fractional coverage over a larger, lower resolution dataset.

A tool to build image composites from Landsat 8 OLI/TIRS datasets. It supports using either GDAL or ArcGIS as its GIS layer and builds composite images from individual bands in a Landsat 8 download. Does not require extracting the downloaded tar.gz file prior to operation.

Build a true color image from a downloaded Landsat 8 OLI/TIRS bundle:

$ python landsattool.py LC80340352014167LGN00.tar.gz visible.tif

Build an image using SWIR for red, NIR for green, and coastal aerosol for blue:

$ python landsattool.py --bands 7 5 1 -- LC80340352014167LGN00.tar.gz visible.tif

The pyrs package contains a collection of Python modules that can be used to automate image processing in other Python applications. The top level modules are built using GIS kernels, the implementations of which are defined in the python.kernel package. The GIS kernels are expected to provide the underlying image I/O and GIS operations that algorithms defined in the top level package may utilize for their analyses.