Satellite imagery for dummies.
TL;DR: Generate JPEG earth imagery from coordinates/location name with publicly available satellite data.
This tool is for a sentient being who wants to view high-res satellite imagery of earth, without digging through all the nitty gritty geospatial details of it. So if this is your first time trying to explore how parts of the Earth look from space, you're at the right place.
NB: felicette at the present state searches for cloud-cover < 10%, and doesn't constrain results on the basis of dates.
One can see Product Roadmap for upcoming features.
felicette depends on GDAL. But the following steps cover GDAL's installation as well.
rio-color uses numpy headers to setup, thus installing numpy and GDAL=={ogrinfo --version} would be sufficient before installing felicette.
$ sudo add-apt-repository ppa:ubuntugis/ppa
$ sudo apt-get update
$ sudo apt-get install python-numpy gdal-bin libgdal-dev
$ gdal-config --version
<version-number>
* activate virtual environment *
$ pip install numpy GDAL==<version-number>
$ pip install felicette
$ brew install gdal
$ gdal-config --version
<version-number>
* activate virtual environment *
$ pip install numpy GDAL==<version-number>
$ pip install felicette
As pointed out here, the following docker image works and is volume-mapped to the present working directory.
$ docker run -it -v "$PWD"/felicette-data:/root/felicette-data milhouse1337/felicette felicette -l "Montreal"
Thanks @milhouse1337 for the docker-image.
rio-color, one of the felicette's dependencies isn't available on conda ecosystem yet. Here's the link to a small discussion on an installation-issue. This section would be updated when there is a stable version of felicette for Windows.
Felicette has plans to build in-house RGB image enhancement algorithms or use imagemagick /[similar tools on conda-forge] for a Windows release, at least until rio-color is available on conda-forge/conda.
To use it:
$ felicette --help
Usage: felicette [OPTIONS]
Satellite imagery for dummies.
Options:
-c, --coordinates FLOAT... Coordinates in (lon, lat) format. This overrides
-l command
-l, --location-name TEXT Location name in string format
-pan, --pan-enhancement Enhance image with panchromatic band
--no-preview Skip previewing of pre-processed low resolution RGB
satellite image.
-v, --vegetation Show Color Infrared image to highlight
vegetation
-V, --version Show the version number and quit
-p, --product TEXT Product name 'landsat'/'sentinel'
--help Show this message and exit.
Felicette can download and process Landsat images taking the location's input as (lon, lat) or the location name. They can be used in the following way.
With location name:
$ felicette -l "Kanyakumari"
With coordinates:
$ felicette -c 77.5385 8.0883
--product / -p option is used to specify which data-product is used to generate images i.e Sentinel or Landsat. By default, Landsat-8 data will be used to generate images.
$ felicette -l "Kanyakumari" -p "sentinel"
NB: *To use sentinel data source, one has to set AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY (To generate a pair, go to AWS console -> My Security Credentials -> Access keys). This is because Sentinel-2 data is in a Requester Pays bucket.
-pan option uses the panchromatic band to enhance image's resolution to 15 meters, contrary to resolution of RGB bands(30 meters) if Landsat product is being used. Felicette doesn't support any panchromatic enhancements for Sentinel-2 data which already have a resolution of 10m.
To get a better Landsat image using felicette use:
$ felicette -pan -c 77.5385 8.0883
--no-preview option doesn't download image to preview, and directly downloads and processes original data. Please use this if you're sure of the location/quality of the images to be generated by felicette with the arguments provided to it.
$ felicette --no-preview -p -c 77.5385 8.0883
-v option generates a CIR image to highlight vegetation in 'red' color. Note that, '-p' option isn't taken into consideration while generating CIR imagery in felicette.
$ felicette -v -l "Kanyakumari"
Félicette was the first cat launched into space, on 18 October 1963. Even though she landed back on earth safely, Félicette was euthanized two months after the launch so that scientists could perform a necropsy to examine her brain. She was the only cat to have survived spaceflight. Here's a footage of the mission from the archives.
When you get a satellite imagery using this tool, imagine Félicette took the picture for you :))
Here are some more sample images generated by felicette.
Here is a link to the original images generated with RGB, CIR options.
Following is a recording of the terminal session recording usage of felicette.
felicette is open-source and welcomes all kinds of contributions, be it documentation/wiki pages/bug-fixes/development of new features.
A small contributors' guide is available here.