def feature_vector(imageSizeX, imageSizeY, peaks, match_pattern: MatchPattern):
'''
This function generates a sparse mask stack to extract a feature vector.
A match template based on the parameters in :code:`parameters` is placed at
each peak position in an individual mask layer. This mask stack can then
be used in :meth:`~libertem.api.Context.create_mask_job` to generate a feature vector for each
frame.
Summing up the mask stack along the first axis generates a mask that can be used for virtual
darkfield imaging of all peaks together.
Parameters
----------
imageSizeX,imageSizeY : int
Frame size in px
peaks : numpy.ndarray
Peak positions in px as numpy.ndarray of shape (n, 2) with integer type
match_pattern : MatchPattern
Instance of :class:`~MatchPattern`
'''
crop_size = match_pattern.get_crop_size()
return masks.sparse_template_multi_stack(
mask_index=range(len(peaks)),
offsetX=peaks[:, 1] - crop_size,
offsetY=peaks[:, 0] - crop_size,
template=match_pattern.get_mask(
(2 * crop_size + 1, 2 * crop_size + 1)),
imageSizeX=imageSizeX,
imageSizeY=imageSizeY,
)
def test_sparse_template_multi_stack():
template = np.ones((2, 3))
stack = m.sparse_template_multi_stack(mask_index=(0, 1, 2),
offsetY=(13, 14, 15),
offsetX=(15, 14, 13),
template=template,
imageSizeY=32,
imageSizeX=32)
t1 = np.zeros((32, 32))
t1[13:15, 15:18] = 1
t2 = np.zeros((32, 32))
t2[14:16, 14:17] = 1
t3 = np.zeros((32, 32))
t3[15:17, 13:16] = 1
assert np.allclose(stack[0].todense(), t1)
assert np.allclose(stack[1].todense(), t2)
assert np.allclose(stack[2].todense(), t3)