EOReader is a remote-sensing opensource python library reading optical and SAR constellations, loading and stacking bands, clouds, DEM and spectral indices in a sensor-agnostic way.

It also implements additional sensor-agnostic features:

• load: Load many band types:
  • satellite bands (optical or SAR)
  • index
  • cloud bands
  • DEM bands
• stack: Stack all these type of bands

EOReader works with xarrays.DataArray and geopandas.GeoDataFrames

from eoreader.reader import Reader
from eoreader.bands import *

# Sentinel-2 path
s2_path = "S2B_MSIL1C_20181126T022319_N0207_R103_T51PWM_20181126T050025.SAFE"

# Create the reader object and open satellite data
reader = Reader()

# The reader will recognize the constellation from its product structure
s2_prod = reader.open(s2_path)

# Load some bands and index
bands = s2_prod.load([NDVI, GREEN, CLOUDS])

# Create a stack with some bands
stack = s2_prod.stack([RED, GREEN, BLUE], stack_path="s2_rgb_stack.tif")

from eoreader.reader import Reader
from eoreader.bands import *

# Sentinel-1 GRD path
s1_path = "S1B_EW_GRDM_1SDH_20200422T080459_20200422T080559_021254_028559_784D.zip"

# Create the reader object and open satellite data
reader = Reader()

# The reader will recognize the constellation from its product structure
s1_prod = reader.open(s1_path)

# Load some bands and index
bands = s1_prod.load([VV, VH])

# Create a stack with some bands
stack = s1_prod.stack([VV_DSPK, VH_DSPK], stack_path="s1_stack.tif")

⚠️SNAP and SAR

SAR products need ESA SNAP free software to be orthorectified and calibrated. Ensure that you have the folder containing your gpt executable in your PATH. If you are using SNAP 8.0, be sure to have your software up-to-date (SNAP version >= 8.0).

The API documentation can be found here.

Available notebooks provided as examples:

• Why EOReader?
• Basic tutorial
• Optical data
• SAR data
• VHR data
• Sentinel-3 data
• Water detection on multiple products
• DEM
• Custom stacks
• Methods to clean optical bands
• S3 Compatible Storage
• Dask
• STAC

You can install EOReader via pip:

pip install eoreader

EOReader mainly relies on geopandas and rasterio (through rioxarray).

On Windows and with pip, you may face installation issues due to GDAL. The well known workaround of installing from Gohlke's wheels also applies here. Please look at the rasterio page to learn more about that.

You can install EOReader via conda:

conda config --env --set channel_priority strict
conda install -c conda-forge eoreader

SERTIT is part of the Copernicus Emergency Management Service rapid mapping and risk and recovery teams.

In these activations, we need to deliver information (such as flood or fire delineations, landslides mapping, etc.) based on various constellations (more than 10 optical and 5 SAR). As every minute counts in production, it seemed crucial to harmonize the ground on which are built our production tools, in order to make them as sensor-agnostic as possible.

Thus, thanks to EOReader, these tools are made independent to the constellation:

• the algorithm (and its developer) can focus on its core tasks (such as extraction) without taking into account the constellation characteristics (how to load a band, which band correspond to which band number, which band to use for this index...)
• the addition of a new constellation is done effortlessly (if existing in EOReader) and without any modification of the algorithm
• the maintenance is simplified and the code is way more readable (no more ifs regarding the sensor type!)
• the testing is also simplified as the sensor-related parts are tested in this library

However, keep in mind that the support of all the constellations used in CEMS is done in the best effort mode, especially for commercial data. Indeed, we may not have faced every product type, sensor mode or order configuration, so some details may be missing. If this happens to you, do not hesitate to make a PR or write an issue about that !

• GeoPython 2022 [ PDF ] [ YouTube ]
• Mentionned in [Live ]SIG 2022 by ESRI France (in French): Enrichir ArcgisPro grâce à des processus personnalisés d'observation de la Terre [ PDF ]

• ESA Success Story

• Maxant, J.; Braun, R.; Caspard, M.; Clandillon, S. ExtractEO, a Pipeline for Disaster Extent Mapping in the Context of Emergency Management. Remote Sens. 2022, 14, 5253. https://doi.org/10.3390/rs14205253 (Technical Note)

EOReader is licensed under Apache License v2.0. See LICENSE file for details.

EOReader has been created by ICube-SERTIT. Follow us on Twitter

EOReader is built on top of amazing libs, without which it couldn't have been coded:

• geopandas
• rasterio
• xarray
• rioxarray