/pansharpening

A commandline tool to process panchromatic and RGB rasters to create a pansharpened image.

Primary LanguagePythonMIT LicenseMIT

Pansharpening tool

A Python processing script which takes the panchromatic and multispectral rasters as an input, and outputs the pansharpened raster.

Setup

  1. If GDAL libraries are not on the system, conda is the simplest way to install them:
$ conda env create -f pansharpening.yml
  • If GDAL libraries are available, skip to step (3), but it's a good idea to create a virtual environment:
$ python -m venv venv && source vevn/bin/activate # on Unix
  1. Activate the newly created environment:
$ conda activate pansharpening
  1. From inside the directory that includes the pyproject.toml file:
$ python -m pip install .

Usage

$ python -m pansharpening -h

usage: pansharpening [-h] [--stream] [--raw] [--file FILE]

Pansharpening pipeline

options:
  -h, --help            show this help message and exit
  --stream, -s          process streamed imagery and discard inputs (default: `False`)
  --compression {DEFLATE,LERC,LERC_DEFLATE,LERC_ZSTD,LZW,PACKBITS,JPEG}, -c {DEFLATE,LERC,LERC_DEFLATE,LERC_ZSTD,LZW,PACKBITS,JPEG}
                        choose the compression algorithm of the raster output (default: `LERC`)
  --raw, -r             derive from raw Pleiades Neo imagery (default: `False`)
  --out OUT, -o OUT     change the output destination directory (default: `./output`)
  --file FILE, -f FILE  TOML file input of source imagery (default: `./data/imagery-sources.toml`)
  --panchromatic PANCHROMATIC, -p PANCHROMATIC
                        Panchromatic image URL
  --rgb RGB, -g RGB     RGB image URL
  • By default, the raster inputs are cached to disk after the first download, but the --stream option changes the processing on-the-fly and discards inputs after finished:
$ python -m pansharpening -s
  • The pansharpened raster is available in the output directory.
  • Change sources file in data/imagery-sources.toml, or provide another sources file using the --file option.
    • Pass in URLs to panchromatic and RGB TIFFs as commandline arguments using the --panchromatic and --rgb options.
  • Change the output directory of the pansharpened raster using the --out option.
  • Alternative compression options are configurable through the --compression flag:
$ python -m pansharpening -c PACKBITS