from sentinelhub import (
DataCollection,
)
from eolearn.io import SentinelHubEvalscriptTask
from eocrops.utils import base_functions as utils
from eolearn.core import linearly_connect_tasks, EOWorkflow, FeatureType, OutputTask
[docs]class CollectionsSH:
def __init__(self, config, resolution=10):
"""
This method aims to download Copernicus products that are defined by SentinelHub collections freely available
Parameters
----------
config : sentinelhub.config
SentinelHub configuration
resolution : int
Spatial resolution of the downloaded products
"""
self.config = config
self.resolution = resolution
[docs] def get_task_PPI(self):
"""Download Productivity and Phenology Index"""
# https://collections.sentinel-hub.com/vegetation-phenology-and-productivity-parameters-season-1/
byoc = DataCollection.define_byoc(
collection_id="67c73156-095d-4f53-8a09-9ddf3848fbb6",
name="PPI",
service_url="https://creodias.sentinel-hub.com",
is_timeless=False,
)
evalscript = """
function setup() {
return {
input: ["SOSD", "EOSD", "LSLOPE", "RSLOPE", "MAXD",
"AMPL", "MAXV", "SPROD", "TPROD",
"QFLAG", "dataMask"],
output: [
{id: "SOSD", bands: 1, sampleType: SampleType.UINT16},
{id: "EOSD", bands: 1, sampleType: SampleType.UINT16},
{id: "LSLOPE", bands: 1, sampleType: SampleType.FLOAT32},
{id: "RSLOPE", bands: 1, sampleType: SampleType.FLOAT32},
{id: "MAXD", bands: 1, sampleType: SampleType.UINT16},
{id: "AMPL", bands: 1, sampleType: SampleType.UINT16},
{id: "MAXV", bands: 1, sampleType: SampleType.FLOAT32},
{id: "SPROD", bands: 1, sampleType: SampleType.UINT16},
{id: "TPROD", bands: 1, sampleType: SampleType.UINT16},
{id: "QFLAG", bands: 1, sampleType: SampleType.UINT16},
{id:"dataMask", bands:1, sampleType: SampleType.UINT16 }]
}
}
function evaluatePixel(sample) {
return {
SOSD: [sample.SOSD],
EOSD: [sample.EOSD],
LSLOPE: [sample.LSLOPE],
RSLOPE: [sample.RSLOPE],
MAXD: [sample.MAXD],
AMPL: [sample.AMPL],
MAXV: [sample.MAXV],
SPROD: [sample.SPROD],
TPROD: [sample.TPROD],
QFLAG: [sample.QFLAG],
dataMask : [sample.dataMask]
}
}
"""
input_task = SentinelHubEvalscriptTask(
features=[
(FeatureType.DATA_TIMELESS, "SOSD", "SOSD"),
(FeatureType.DATA_TIMELESS, "EOSD", "EOSD"),
(FeatureType.DATA_TIMELESS, "RSLOPE", "RSLOPE"),
(FeatureType.DATA_TIMELESS, "LSLOPE", "LSLOPE"),
(FeatureType.DATA_TIMELESS, "MAXD", "MAXD"),
(FeatureType.DATA_TIMELESS, "AMPL", "AMPL"),
(FeatureType.DATA_TIMELESS, "MAXV", "MAXV"),
(FeatureType.DATA_TIMELESS, "SPROD", "SPROD"),
(FeatureType.DATA_TIMELESS, "TPROD", "TPROD"),
(FeatureType.DATA_TIMELESS, "QFLAG", "QFLAG"),
(FeatureType.DATA_TIMELESS, "dataMask", "dataMask"),
],
data_collection=DataCollection.PPI,
evalscript=evalscript,
resolution=self.resolution,
config=self.config,
)
return input_task
[docs] def get_task_WC(self):
"""Download World Cover (10m) : only for 2020-2021"""
# https://collections.sentinel-hub.com/worldcover/
byoc = DataCollection.define_byoc(
collection_id="0b940c63-45dd-4e6b-8019-c3660b81b884",
name="WC",
service_url=" http://services.sentinel-hub.com",
is_timeless=False,
)
evalscript = """
function setup() {
return {
input: ["Map", "dataMask"],
output: [
{id: "WC", bands: 1, sampleType: SampleType.FLOAT32},
{id:"dataMask", bands:1, sampleType: SampleType.UINT8 }
]
}
}
function evaluatePixel(sample) {
return {
WC: [sample.Map],
dataMask : [sample.dataMask]
}
}
"""
# https://collections.sentinel-hub.com/vegetation-phenology-and-productivity-parameters-season-1/
input_task = SentinelHubEvalscriptTask(
features=[(FeatureType.DATA, "WC"), (FeatureType.MASK, "dataMask")],
data_collection=byoc,
resolution=self.resolution,
config=self.config,
evalscript=evalscript,
)
return input_task
[docs] def get_task_GLC(self):
"""Download Global Land Cover (100m)"""
# 100 m Land Cover products for the years 2015 - 2019
# https://collections.sentinel-hub.com/worldcover/
byoc = DataCollection.define_byoc(
collection_id="f0a97620-0e88-4c1f-a1ac-bb388fabdf2c",
name="GLC",
service_url="https://creodias.sentinel-hub.com",
is_timeless=False,
)
evalscript = """
function setup() {
return {
input: ["Discrete_Classification_map"],
output: [
{id: "GLC", bands: 1, sampleType: SampleType.FLOAT32},
{id:"dataMask", bands:1, sampleType: SampleType.UINT8 }
]
}
}
function evaluatePixel(sample) {
return {
GLC: [sample.Discrete_Classification_map],
dataMask : [sample.dataMask]
}
}
"""
# https://collections.sentinel-hub.com/vegetation-phenology-and-productivity-parameters-season-1/
input_task = SentinelHubEvalscriptTask(
features=[(FeatureType.DATA, "GLC"), (FeatureType.MASK, "dataMask")],
data_collection=DataCollection.GLC,
resolution=self.resolution,
config=self.config,
evalscript=evalscript,
)
return input_task
def execute(self, input_task, shapefile, time_stamp):
"""
For a given input task using get methods, get the pixel informations into EOPatch
Parameters
----------
input_task : SentinelHubEvalscriptTask
Input task for the collection
shapefile : gpd.GeoDataFrame
Shapefile containing the field boundaries
time_stamp : tuple
Time periods, if it gathers two years, you would have two timestamps
Returns
-------
EOPatch
"""
field_bbox = get_bounding_box(shapefile)
# With eolearn 1.0
output_task = OutputTask("eopatch")
workflow_nodes = linearly_connect_tasks(input_task, output_task)
workflow = EOWorkflow(workflow_nodes)
result = workflow.execute(
{workflow_nodes[0]: {"bbox": field_bbox, "time_interval": time_stamp}}
)
return result.outputs["eopatch"]
