extract_ROI

museotoolbox.processing.extract_ROI(in_image, in_vector, *fields, **kwargs)[source]

Extract raster values from Regions Of Interest in a vector file.

Initially written by Mathieu Fauvel, improved by Nicolas Karasiak.

Parameters

  • in_image (str.) – the name or path of the raster file, could be any file that GDAL can open.

  • in_vector (str.) – A filename or path corresponding to a vector file. It could be any file that GDAL/OGR can open.

  • *fields (str.) – Each field to extract label/value from.

  • **kwargs (list of kwargs.) –

    • get_pixel_position : bool, optional (default=False).

    If get_pixel_position=True, will return pixel position in the image for each point.

    • only_pixel_position : bool, optional (default=False).

    If only_pixel_position=True, with only return pixel position for each point.

    • prefer_memory : bool, optional (default=True).

    If prefer_memory=False, will write temporary raster on disk to extract ROI values.

    • verbose : bool or int, optional (default=True).

    The higher is the int verbose, the more it will returns informations.

Returns

  • X (np.ndarray, size of (n_samples,n_features).) – The sample matrix. A n*d matrix, where n is the number of referenced pixels and d is the number of features. Each line of the matrix is a pixel.

  • y (np.ndarray, size of (n_samples,).) – The label of each pixel.

See also

museotoolbox.processing.read_vector_values()

read field values from vector file.

Examples

>>> from museotoolbox.datasets import load_historical_data
>>> from museotoolbox.processing import extract_ROI
>>> raster,vector= load_historical_data()
>>> X,Y = extract_ROI(raster,vector,'Class')
>>> X
array([[ 213.,  189.,  151.],
   [ 223.,  198.,  158.],
   [ 212.,  188.,  150.],
   ...,
   [ 144.,  140.,  105.],
   [  95.,   92.,   57.],
   [ 141.,  137.,  102.]])
>>> X.shape
(12647,3)
>>> Y
[3 3 3 ..., 1 1 1]
>>> Y.shape
(12647,)