Custom stacks#
Let’s use EOReader with custom stacks.
# EOReader Imports
import os
from eoreader.reader import Reader
from eoreader.products import SensorType
from eoreader.bands import BLUE, GREEN, RED, NIR, SWIR_1, VV, VV_DSPK, SLOPE, HILLSHADE
from sertit import display
reader = Reader()
# Create logger
import logging
from sertit import logs
logger = logging.getLogger("eoreader")
logs.init_logger(logger)
# Set a DEM
from eoreader.env_vars import DEM_PATH
os.environ[DEM_PATH] = os.path.join("/home", "ds2_db2", "BASES_DE_DONNEES", "GLOBAL", "COPDEM_30m",
"COPDEM_30m.vrt")
Custom stack with minimum data#
For both SAR and optical stacks, the two minimum keywords to provide are:
band_map: a dictionary mapping the satellite band to the band number (starting to 1, in GDAL style)sensor_type: EitherSARorOPTICAL(a string or a SensorType Enum)
# Paths
stack_folder = os.path.join("/home", "ds2_db3", "CI", "eoreader", "others")
opt_path = os.path.join(stack_folder, "20200310T030415_WV02_Ortho_BGRN_STK.tif")
sar_path = os.path.join(stack_folder, "20210827T162210_ICEYE_SC_GRD_STK.tif")
# Optical minimum example
opt_prod = reader.open(opt_path,
custom=True,
sensor_type="OPTICAL", # With a string
band_map={BLUE: 1, GREEN: 2, RED: 3, NIR: 4, SWIR_1: 5})
opt_prod
eoreader.CustomProduct '20200310T030415_WV02_Ortho_BGRN_STK'
Attributes:
condensed_name: 20200310T030415_WV02_Ortho_BGRN_STK
path: /home/ds2_db3/CI/eoreader/others/20200310T030415_WV02_Ortho_BGRN_STK.tif
constellation: CUSTOM
sensor type: Optical
product type: CUSTOM
default pixel size: 8.0
default resolution: None
acquisition datetime: 2025-12-23T11:37:05.519940
band mapping:
BLUE: 1
GREEN: 2
RED: 3
NIR: 4
SWIR_1: 5
needs extraction: False
opt_stack = opt_prod.stack([BLUE, GREEN, RED])
2025-12-23 11:37:05,528 - [DEBUG] - Loading bands ['BLUE', 'GREEN', 'RED']
/opt/conda/lib/python3.11/site-packages/rasterio/warp.py:387: NotGeoreferencedWarning: Dataset has no geotransform, gcps, or rpcs. The identity matrix will be returned.
dest = _reproject(
2025-12-23 11:37:09,094 - [DEBUG] - Stacking
opt_stack[:, ::5, ::5].plot.imshow(robust=True)
<matplotlib.image.AxesImage at 0x7f55bc1aea90>
opt_stack
<xarray.DataArray 'BLUE_GREEN_RED' (bands: 3, y: 2237, x: 1244)> Size: 33MB
dask.array<where, shape=(3, 2237, 1244), dtype=float32, chunksize=(1, 1024, 1024), chunktype=numpy.ndarray>
Coordinates:
* x (x) float64 10kB 3.044e+05 3.044e+05 ... 3.143e+05 3.143e+05
* y (y) float64 18kB 1.459e+06 1.459e+06 ... 1.441e+06 1.441e+06
spatial_ref int64 8B 0
* bands (bands) object 24B MultiIndex
* variable (bands) object 24B SpectralBandNames.BLUE ... SpectralBandNa...
* band (bands) int64 24B 1 1 1
Attributes:
path: /tmp/tmpbxzv8a7q/tmp_20200310T030415_WV02_Ortho_BGRN_S...
long_name: BLUE GREEN RED
constellation: CUSTOM
constellation_id: CUSTOM
product_path: /home/ds2_db3/CI/eoreader/others/20200310T030415_WV02_...
product_name: 20200310T030415_WV02_Ortho_BGRN_STK
product_filename: 20200310T030415_WV02_Ortho_BGRN_STK
instrument: CUSTOM
product_type: CUSTOM
acquisition_date: 20251223T113709
condensed_name: 20200310T030415_WV02_Ortho_BGRN_STK
orbit_direction: None# SAR minimum example
sar_prod = reader.open(
sar_path,
custom=True,
sensor_type=SensorType.SAR, # With the Enum
band_map={VV: 1, VV_DSPK: 2}
)
sar_prod
eoreader.CustomProduct '20210827T162210_ICEYE_SC_GRD_STK'
Attributes:
condensed_name: 20210827T162210_ICEYE_SC_GRD_STK
path: /home/ds2_db3/CI/eoreader/others/20210827T162210_ICEYE_SC_GRD_STK.tif
constellation: CUSTOM
sensor type: SAR
product type: CUSTOM
default pixel size: 48.0
default resolution: None
acquisition datetime: 2025-12-23T11:37:09.516545
band mapping:
VV: 1
VV_DSPK: 2
needs extraction: False
sar_stack = sar_prod.stack([SLOPE, VV, VV_DSPK])
2025-12-23 11:37:09,527 - [DEBUG] - Loading bands ['VV', 'VV_DSPK']
2025-12-23 11:37:12,503 - [DEBUG] - Loading DEM bands ['SLOPE']
2025-12-23 11:37:12,504 - [DEBUG] - Warping DEM for 20210827T162210_ICEYE_SC_GRD_STK
2025-12-23 11:37:12,507 - [DEBUG] - Using DEM: /home/ds2_db2/BASES_DE_DONNEES/GLOBAL/COPDEM_30m/COPDEM_30m.vrt
2025-12-23 11:37:15,395 - [DEBUG] - Computing slope for 20210827T162210_ICEYE_SC_GRD_STK
2025-12-23 11:37:23,950 - [DEBUG] - Stacking
sar_stack[:, ::5, ::5].plot.imshow(robust=True)
<matplotlib.image.AxesImage at 0x7f557e8e6350>
sar_stack
<xarray.DataArray 'SLOPE_VV_VV_DSPK' (bands: 3, y: 2748, x: 2967)> Size: 98MB
dask.array<where, shape=(3, 2748, 2967), dtype=float32, chunksize=(1, 1024, 1024), chunktype=numpy.ndarray>
Coordinates:
spatial_ref int64 8B 0
* y (y) float64 22kB 1.113e+04 1.109e+04 ... -1.206e+05 -1.207e+05
* x (x) float64 24kB 6.7e+05 6.701e+05 ... 8.124e+05 8.124e+05
* bands (bands) object 24B MultiIndex
* variable (bands) object 24B DemBandNames.SLOPE ... SarBandNames.VV_DSPK
* band (bands) int64 24B 1 1 1
Attributes:
path: /tmp/tmprnkgwy83/tmp_20210827T162210_ICEYE_SC_GRD_STK/...
long_name: SLOPE VV VV_DSPK
constellation: CUSTOM
constellation_id: CUSTOM
product_path: /home/ds2_db3/CI/eoreader/others/20210827T162210_ICEYE...
product_name: 20210827T162210_ICEYE_SC_GRD_STK
product_filename: 20210827T162210_ICEYE_SC_GRD_STK
instrument: CUSTOM
product_type: CUSTOM
acquisition_date: 20251223T113723
condensed_name: 20210827T162210_ICEYE_SC_GRD_STK
orbit_direction: None# You can compute the footprint and the extent
# Optical
extent = opt_prod.extent()
footprint = opt_prod.footprint()
base = extent.plot(color='cyan', edgecolor='black')
footprint.plot(ax=base, color='blue', edgecolor='black', alpha=0.5)
base.set_title("Optical footprint and extent");
'_vectorize' function is not lazy yet. Computing the raster.
# SAR geometry
extent = sar_prod.extent()
footprint = sar_prod.footprint()
base = extent.plot(color='cyan', edgecolor='black')
footprint.plot(ax=base, color='blue', edgecolor='black', alpha=0.5)
base.set_title("SAR footprint and extent");
'_vectorize' function is not lazy yet. Computing the raster.
Custom stack with full data#
If you know them, it is best to give EOReader all the data you know about your stack:
name: product name. If not provided, the filename will be useddatetime: product acquisition datetime. If not provided, the datetime of the creation of the object will be usedconstellation: product constellation. If not provided,CUSTOMwill be set. Either a string of a Constellation enum.product_type: product type. If not provided,CUSTOMwill be set.pixel_size: product default pixel size. If not provided, the stack pixel size will be used.
For optical products, two additional keyword can be set to compute the hillshade band:
sun_azimuthsun_zenith
# Optical
opt_prod = reader.open(
opt_path,
custom=True,
name="20200310T030415_WV02_Ortho",
datetime="20200310T030415",
sensor_type=SensorType.OPTICAL,
constellation="WV02",
product_type="Ortho",
pixel_size=2.0,
sun_azimuth=10.0,
sun_zenith=20.0,
band_map={BLUE: 1, GREEN: 2, RED: 3, NIR: 4, SWIR_1: 5},
)
hillshade = opt_prod.load(HILLSHADE)[HILLSHADE]
2025-12-23 11:37:26,662 - [DEBUG] - Loading DEM bands ['HILLSHADE']
2025-12-23 11:37:26,663 - [DEBUG] - Warping DEM for 20200310T030415_WV02_Ortho
2025-12-23 11:37:26,665 - [DEBUG] - Using DEM: /home/ds2_db2/BASES_DE_DONNEES/GLOBAL/COPDEM_30m/COPDEM_30m.vrt
2025-12-23 11:37:29,582 - [DEBUG] - Computing hillshade DEM for 20200310T030415_WV02_Ortho
hillshade.plot(robust=True, cmap="mako")
<matplotlib.collections.QuadMesh at 0x7f557e7ee990>
# SAR
sar_prod = reader.open(
sar_path,
custom=True,
sensor_type=SensorType.SAR,
name="20210827T162210_ICEYE_SC_GRD",
datetime="20210827T162210",
constellation="ICEYE",
product_type="GRD",
pixel_size=6.0,
band_map={VV: 1, VV_DSPK: 2},
)
from pprint import pprint
from eoreader import utils
# Read and display metadata
mtd, _ = sar_prod.read_mtd()
pprint(utils.quick_xml_to_dict(mtd))
('custom_metadata',
{'band_map': "{'VV': 1, 'VV_DSPK': 2}",
'cloud_cover': 'None',
'condensed_name': '20210827T162210_ICEYE_GRD',
'constellation': 'ICEYE',
'datetime': '2021-08-27T16:22:10',
'id': 'None',
'instrument': 'CUSTOM',
'name': '20210827T162210_ICEYE_SC_GRD',
'orbit_direction': 'None',
'pixel_size': '6.0',
'product_type': 'GRD',
'sensor_type': 'SAR',
'sun_azimuth': 'None',
'sun_zenith': 'None'})
