datacube.api.utils module

class datacube.api.utils.PqaMask

Bases: enum.Enum

class datacube.api.utils.WofsMask

Bases: enum.Enum

class datacube.api.utils.OutputFormat

Bases: enum.Enum

class datacube.api.utils.PercentileInterpolation

Bases: enum.Enum

datacube.api.utils.empty_array(shape, dtype=<type 'numpy.int16'>, fill=-999)

Return an empty (i.e. filled with the no data value) array of the given shape and data type

Parameters:

• shape – shape of the array
• dtype – data type of the array (defaults to int32)
• fill – no data value (defaults to -999)

Returns:

array

class datacube.api.utils.DatasetBandMetaData(no_data_value, data_type)

no_data_value = None

data_type = None

class datacube.api.utils.DatasetMetaData(shape, transform, projection, bands)

shape = None

transform = None

projection = None

bands = None

pixel_size_x = None

pixel_size_y = None

ul = None

lr = None

datacube.api.utils.get_dataset_metadata(dataset)

datacube.api.utils.get_dataset_data(dataset, bands=None, x=0, y=0, x_size=None, y_size=None)

datacube.api.utils.read_dataset_data(dataset, bands=None, x=0, y=0, x_size=None, y_size=None)

Return one or more bands from a dataset

Parameters:

• dataset – The dataset from which to read the band
• bands – A list of bands to read from the dataset
• x –
• y –
• x_size –
• y_size –

Returns:

dictionary of band/data as numpy array

datacube.api.utils.get_dataset_data_masked(dataset, bands=None, x=0, y=0, x_size=None, y_size=None, ndv=-999, mask=None)

Return one or more bands from the dataset with pixel quality applied

Return type:dict[numpy.array]

datacube.api.utils.get_dataset_data_with_pq(dataset, dataset_pqa, bands=None, x=0, y=0, x_size=None, y_size=None, masks_pqa=[<PqaMask.PQ_MASK_CLEAR: 16383>], ndv=-999)

Return one or more bands from the dataset with pixel quality applied

Return type:dict[numpy.array]

datacube.api.utils.apply_mask(data, mask, ndv=-999)

datacube.api.utils.get_mask_pqa(pqa, pqa_masks=[<PqaMask.PQ_MASK_CLEAR: 16383>], x=0, y=0, x_size=None, y_size=None, mask=None)

Return a pixel quality mask

Parameters:

• pqa – Pixel Quality dataset
• pqa_masks – which PQ flags to use
• mask – an optional existing mask to update

Returns:

the mask

datacube.api.utils.consolidate_masks(masks)

datacube.api.utils.get_mask_wofs(wofs, wofs_masks=[<WofsMask.WET: 128>], x=0, y=0, x_size=None, y_size=None, mask=None)

Return a WOFS mask

Parameters:

• wofs – WOFS dataset
• wofs_masks – which WOFS values to mask
• mask – an optional existing mask to update

Returns:

the mask

datacube.api.utils.get_mask_vector_for_cell(x, y, vector_file, vector_layer, vector_feature, width=4000, height=4000, pixel_size_x=0.00025, pixel_size_y=-0.00025)

Return a mask for the given cell based on the specified feature in the vector file

Parameters:

• x (int) – X cell index
• y (int) – X cell
• vector_file (str) – Vector file containing the mask polygon
• vector_layer (str) – Layer name within the vector file
• vector_feature (int) – Feature id (index starts at 0) within the layer
• width (int) – Width of the mask
• height (int) – Height of the mask
• pixel_size_x (float) – X pixel size
• pixel_size_y (float) – Y pixel size

Returns:

The mask

Return type:

numpy.ma.MaskedArray.mask (array of boolean)

datacube.api.utils.get_dataset_data_stack(tiles, dataset_type, band_name, x=0, y=0, x_size=None, y_size=None, ndv=None, mask_pqa_apply=False, mask_pqa_mask=None)

datacube.api.utils.raster_create(path, data, transform, projection, no_data_value, data_type, options=['INTERLEAVE=PIXEL'], width=None, height=None, dataset_metadata=None, band_ids=None)

datacube.api.utils.raster_create_geotiff(path, data, transform, projection, no_data_value, data_type, options=['INTERLEAVE=PIXEL'], width=None, height=None, dataset_metadata=None, band_ids=None)

Create a raster from a list of numpy arrays

Parameters:

• path – path to the output raster
• data – list of numpy arrays
• transform – geo transform
• projection – projection
• no_data_value – no data value
• data_type – data type
• options – raster creation options

datacube.api.utils.raster_create_envi(path, data, transform, projection, no_data_value, data_type, options=['INTERLEAVE=BSQ'], width=None, height=None, dataset_metadata=None, band_ids=None)

Create a raster from a list of numpy arrays

Parameters:

• path – path to the output raster
• data – list of numpy arrays
• transform – geo transform
• projection – projection
• no_data_value – no data value
• data_type – data type
• options – raster creation options

datacube.api.utils.propagate_using_selected_pixel(a, b, c, d, ndv=-999)

datacube.api.utils.calculate_ndvi(red, nir, input_ndv=-999, output_ndv=-999)

Calculate the Normalised Difference Vegetation Index (NDVI) from a Landsat dataset

NDVI is defined as (NIR - RED) / (NIR + RED)

datacube.api.utils.calculate_evi(red, blue, nir, l=1, c1=6, c2=7.5, input_ndv=-999, output_ndv=-999)

Calculate the Enhanced Vegetation Index (EVI) from a Landsat dataset first applying Pixel Quality indicators

EVI is defined as 2.5 * (NIR - RED) / (NIR + C1 * RED - C2 * BLUE + L)

Defaults to the standard MODIS EVI of L=1 C1=6 C2=7.5

datacube.api.utils.calculate_nbr(nir, swir, input_ndv=-999, output_ndv=-999)

Calculate the Normalised Burn Ratio (NBR) from a Landsat dataset

NBR is defined as (NIR - SWIR 2) / (NIR + SWIR 2)

class datacube.api.utils.TasselCapIndex

Bases: enum.Enum

datacube.api.utils.calculate_tassel_cap_index(bands, coefficients, input_ndv=-999, output_ndv=nan)

Parameters:

• bands –
• coefficients –
• input_ndv –
• output_ndv –

Returns:

datacube.api.utils.calculate_medoid(X, dist=None)

datacube.api.utils.latlon_to_xy(lat, lon, transform)

Convert lat/lon to x/y for raster NOTE: No projection done - assumes raster has native lat/lon projection

Parameters:

• lat – latitude
• lon – longitude
• transform – GDAL GeoTransform

Returns:

x, y pair

datacube.api.utils.latlon_to_cell(lat, lon)

Return the cell that contains the given lat/lon pair

NOTE: x of cell represents min (contained) lon value but y of cell represents max (not contained) lat value

that is, 120/-20 contains lon values 120->120.99999 but lat values -19->-19.99999 that is, that is, 120/-20 does NOT contain lat value of -20

Parameters:

• lat – latitude
• lon – longitude

Returns:

cell as x, y pair

datacube.api.utils.extract_fields_from_filename(filename)

Parameters:filename – Returns:

datacube.api.utils.intersection(a, b)

datacube.api.utils.union(a, b)

datacube.api.utils.subset(a, b)

datacube.api.utils.get_satellite_string(satellites)

datacube.api.utils.check_overwrite_remove_or_fail(path, overwrite)

datacube.api.utils.log_mem(s=None)

datacube.api.utils.date_to_integer(d)

datacube.api.utils.get_dataset_filename(dataset, output_format=<OutputFormat.GEOTIFF: 'GTiff'>, mask_pqa_apply=False, mask_wofs_apply=False, mask_vector_apply=False)

datacube.api.utils.get_dataset_band_stack_filename(dataset, band, output_format=<OutputFormat.GEOTIFF: 'GTiff'>, mask_pqa_apply=False, mask_wofs_apply=False, mask_vector_apply=False)

datacube.api.utils.get_dataset_datatype(dataset)

datacube.api.utils.get_dataset_type_datatype(dataset_type)

datacube.api.utils.get_dataset_ndv(dataset)

datacube.api.utils.get_dataset_type_ndv(dataset_type)

datacube.api.utils.get_band_name_union(dataset_type, satellites)

datacube.api.utils.get_band_name_intersection(dataset_type, satellites)

datacube.api.utils.format_date(d)

datacube.api.utils.format_date_time(d)

datacube.api.utils.extract_feature_geometry_wkb(vector_file, vector_layer=0, vector_feature=0, epsg=4326)

datacube.api.utils.maskify_stack(stack, ndv=-999)

datacube.api.utils.calculate_stack_statistic_count(stack, ndv=-999, dtype=<type 'numpy.int16'>)

datacube.api.utils.calculate_stack_statistic_count_observed(stack, ndv=-999, dtype=<type 'numpy.int16'>)

datacube.api.utils.calculate_stack_statistic_min(stack, ndv=-999, dtype=<type 'numpy.int16'>)

datacube.api.utils.calculate_stack_statistic_max(stack, ndv=-999, dtype=<type 'numpy.int16'>)

datacube.api.utils.calculate_stack_statistic_mean(stack, ndv=-999, dtype=<type 'numpy.int16'>)

datacube.api.utils.calculate_stack_statistic_percentile(stack, percentile, interpolation=<PercentileInterpolation.NEAREST: 'nearest'>, ndv=-999, dtype=<type 'numpy.int16'>)

datacube.api.utils.get_mask_aoi_cell(vector_file, vector_layer, vector_feature, x, y, width=4000, height=4000, epsg=4326)

datacube.api.utils.grand_mean(means)

datacube.api.utils.combine_means(mean1, count1, mean2, count2)