Textory package

textory.textures module

textory.textures.cross_variogram(x, y, lag=1, win_size=5, win_geom='square', **kwargs)[source]

Calculate moveing window pseudo-variogram with specified lag for the two arrays.

Parameters:
  • y (x,) – Input array
  • lag (int) – Lag distance for variogram, defaults to 1.
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window.
  • win_geom ({"square", "round"}) – Geometry of the kernel. Defaults to square.
Returns:

array like – Array where each element is the pseudo-variogram between the two arrays of the window around the element.
textory.textures.madogram(x, lag=1, win_size=5, win_geom='square', **kwargs)[source]

Calculate moveing window madogram with specified lag for array.

Parameters:
  • x (array like) – Input array
  • lag (int) – Lag distance for variogram, defaults to 1.
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window.
  • win_geom ({"square", "round"}) – Geometry of the kernel. Defaults to square.
Returns:

array like – Array where each element is the madogram of the window around the element
textory.textures.pseudo_cross_variogram(x, y, lag=1, win_size=5, win_geom='square', **kwargs)[source]

Calculate moveing window pseudo-variogram with specified lag for the two arrays.

Parameters:
  • y (x,) – Input array
  • lag (int) – Lag distance for variogram, defaults to 1.
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window.
  • win_geom ({"square", "round"}) – Geometry of the kernel. Defaults to square.
Returns:

array like – Array where each element is the pseudo-variogram between the two arrays of the window around the element.
textory.textures.rodogram(x, lag=1, win_size=5, win_geom='square', **kwargs)[source]

Calculate moveing window rodogram with specified lag for array.

Parameters:
  • x (array like) – Input array
  • lag (int) – Lag distance for variogram, defaults to 1.
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window.
  • win_geom ({"square", "round"}) – Geometry of the kernel. Defaults to square.
Returns:

array like – Array where each element is the madogram of the window around the element
textory.textures.tpi(x, win_size=5, win_geom='square', **kwargs)[source]

Calculate topographic position index for a given window size.

Parameters:
  • x (array like) – Input array
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window. Defaults to 5.
  • win_geom ({"square", "round"}) – Geometry of the kernel. Defaults to square.
Returns:

array like – Array with tpi
textory.textures.variogram(x, lag=1, win_size=5, win_geom='square', **kwargs)[source]

Calculate moveing window variogram with specified lag for array.

Parameters:
  • x (array like) – Input array
  • lag (int) – Lag distance for variogram, defaults to 1.
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window.
  • win_geom ({"square", "round"}) – Geometry of the kernel. Defaults to square.
Returns:

array like – Array where each element is the variogram of the window around the element
textory.textures.window_statistic(x, stat='nanmean', win_size=5, **kwargs)[source]

Calculate the specified statistic with a moveing window of size win_size.

Parameters:
  • x (array like) – Input array
  • stat ({"nanmean", "nanmax", "nanmin", "nanmedian", "nanstd"}) – Statistical measure to calculate.
  • win_size (int, optional) – Length of one side of window. Window will be of size window*window.
  • kwargs (optional) – Any parameters a certain stat may need other than the array itself.
Returns:

array like

textory.statistics module

textory.statistics.pseudo_cross_variogram(x, y, lag=1)[source]

Calculate pseudo-variogram with specified lag for the two arrays.

Parameters:
  • y (x,) – Input arrays
  • lag (int) – Lag distance for variogram, defaults to 1.
Returns:

float – Pseudo-variogram between the two arrays
textory.statistics.variogram(x, lag=1)[source]

Calculate variogram with specified lag for array.

Parameters:
  • x (array like) – Input array
  • lag (int) – Lag distance for variogram, defaults to 1.
Returns:

float – Variogram

textory.wrappers module

Textory satpy Scene and xarray Dataset wrappers

Calculate textures for Scene

With the textures_for_scene() function it is easy to calculate one or multiple textures with the same or different parameters for datasets in a satpy.scene.Scene. The function will return a new Scene either with the textures in addition to all input datasets (default) or a Scene only with the textures, depending on the append parameter of the function.

The textures parameter takes a dictionary where the keys are a tuple with the texture and the parameters which to calculate and the values are a list (or list of tuples in the case of textures which require two inputs) of the datasets to apply the texture to in the general form of:

textures_dict = {("texture_name", lag, win_size, win_geom): [list of dataset names]}

The following example would calculate the variogram with lag=2, win_size=7, win_geom=”square” for the datasets with name “IR_039” and “IR_108” as well as the cross variogram with lag=1, win_size=5, and win_geom=”round” between the datasets with name “WV_062” and “IR_108” of the Scene.

import textory as tx

scn = Scene(...)
textures_dict = {("variogram", 2, 7, "square"): ["IR_039", "IR_108"],
                 ("cross_variogram", 1, 5, "round"): [("WV_062", "IR_108")]}
scn_with_textures = tx.textures_for_scene(scn, textures=textures_dict)

Calculate textures for xarray Dataset

The textures_for_xr_dataset() function works similarly to the textures_for_scene() function above but takes xarray.Dataset as input and also returns a xarray.Dataset.

textory.wrappers.textures_for_scene(scn, textures, append=True)[source]

Wrapper to calculate multiple textures for datasets in a satpy.scene.Scene.

Parameters:
  • scn (satpy.scene.Scene) –
  • textures (dict) –

    Dictionary with textures bands to calulate. The accepted notation is {(“texture name”, lag, win_size, win_geom): [list of dataset names or tuples of dataset names]}

    For example:
    {(“variogram”, 2, 7, “round”), [“IR_039”, “IR_103”]}
  • append (boolean, optional) – If False returns a new satpy.scene.Scene with all calculated textures, By default returns a new satpy.scene.Scene with all input datasets and all calculated textures.
Returns:

satpy.scene.Scene
textory.wrappers.textures_for_xr_dataset(xrds, textures, append=True)[source]

Wrapper to calculate multiple textures for dataarrays in a xarray.Dataset.

Parameters:
  • xrds (xarray.Dataset) –
  • textures (dict) –

    Dictionary with textures bands to calulate. The accepted notation is {(“texture name”, lag, win_size, win_geom): [list of dataset names or tuples of dataset names]}

    For example:
    {(“variogram”, 2, 7, “round”), [“IR_039”, “IR_103”]}
  • append (boolean, optional) – If False returns a new xarray.Dataset with all calculated textures, By default returns a new xarray.Dataset with all input datasets and all calculated textures.
Returns:

xarray.Dataset – Dataset