def remove_large_objects(ar, max_size=128, connectivity=1):
if max_size == -1: # shortcut for efficiency
return ar
out = ar.copy()
if out.dtype == bool:
selem = generate_binary_structure(ar.ndim, connectivity)
ccs = np.zeros_like(ar, dtype=np.int32)
label_image(ar, selem, output=ccs)
else:
ccs = out
try:
component_sizes = np.bincount(ccs.ravel())
except ValueError:
raise ValueError("Negative value labels are not supported. Try "
"relabeling the input with `scipy.ndimage.label` or "
"`skimage.morphology.label`.")
too_big = component_sizes > max_size
too_big_mask = too_big[ccs]
out[too_big_mask] = 0
return out
def find_big_ones_clusters(bin_mask,
min_cluster_length=None,
min_cluster_order=None):
labeled_mask, _ = label_image(bin_mask)
cluster_labels, cluster_sizes = np.unique(labeled_mask,
return_counts=True)
# sort from max to min
sorted_indexes = np.flip(cluster_sizes.argsort())
cluster_sizes = cluster_sizes[sorted_indexes]
cluster_labels = cluster_labels[sorted_indexes]
min_cluster_length = cluster_sizes[min_cluster_order] if min_cluster_order else min_cluster_length
big_clusters_indexes = cluster_sizes > min_cluster_length
big_ones_cluster_labels = cluster_labels[big_clusters_indexes]
# TODO:refactor this part
contour_mask = np.zeros(bin_mask.shape, dtype=bool)
for label in big_ones_cluster_labels:
if label == 0:
continue
contour_mask = np.logical_or(contour_mask, _create_mask_layer_for_label(labeled_mask, label))
return contour_mask
def find_mask_longest_contours(bin_img_with_contours,
filter_by_contour_length=False,
max_number_of_contours=None,
min_contour_length=None):
labeled_img, _ = label_image(bin_img_with_contours)
unique_labels, unique_counts = np.unique(labeled_img, return_counts=True)
longest_contours_indexes = np.flip(unique_counts.argsort())
if not filter_by_contour_length:
if len(longest_contours_indexes) < max_number_of_contours:
max_number_of_contours = len(longest_contours_indexes)
longest_contours_labels = unique_labels[longest_contours_indexes][
0:max_number_of_contours]
else:
longest_contours_lens = unique_counts[longest_contours_indexes][
0:max_number_of_contours]
longest_contours_lens_lower_max = longest_contours_lens > min_contour_length
longest_contours_indexes_sampled = longest_contours_indexes[
longest_contours_lens_lower_max]
longest_contours_labels = unique_labels[
longest_contours_indexes_sampled]
# print("number of selected contour: ", len(longest_contours_labels))
contour_mask = np.zeros(labeled_img.shape, dtype=bool)
for label in longest_contours_labels:
if label == 0:
continue
contour_mask = np.logical_or(
contour_mask, _create_mask_layer_for_label(labeled_img, label))
return contour_mask
def _apply_convective_criterion4(convective_flag_matrix, min_size_pixels):
"""Applies criterion 4 for convective classification.
Criterion 4 states: if pixel (i, j) is marked convective but is not part of
a connected region with size of >= K pixels, (i, j) is not actually
convective.
:param convective_flag_matrix: M-by-N numpy array of Boolean flags (True
if convective, False if not).
:param min_size_pixels: Minimum size of connected region.
:return: convective_flag_matrix: Updated version of input.
"""
region_id_matrix = label_image(convective_flag_matrix.astype(int),
structure=numpy.full((3, 3), 1.))[0]
num_regions = numpy.max(region_id_matrix)
for i in range(num_regions):
these_indices = numpy.where(region_id_matrix == i + 1)
if len(these_indices[0]) >= min_size_pixels:
continue
convective_flag_matrix[these_indices] = False
return convective_flag_matrix
def get_pore_volume_distribution(levitatting_volume, structure_neighbors_num):
structure = get_structure(neighbors_num=structure_neighbors_num)
if levitatting_volume.ndim == 2:
structure = structure[1]
connected_components, _ = label_image(levitatting_volume, structure)
pore_volume_distribution = np.unique(connected_components,
return_counts=True)[1][1:]
return pore_volume_distribution
def get_sparse_masks(
segmented_img: np.ndarray,
raw_img_shape: tuple,
edge_method: str,
selem: int,
transpose=True
) -> np.ndarray:
# allocate array with same size as the raw input image
# so that the dims match for CNMF
img = np.zeros(raw_img_shape, dtype=segmented_img.dtype)
# fill the allocated array with vals from the NuSeT segmentation
# sometimes the segmented image has its dims trimmed by a few pixels
if len(raw_img_shape) == 3:
img[:, :segmented_img.shape[1], :segmented_img.shape[2]] = segmented_img
struc = np.array(
[[[0, 0, 0],
[0, 0, 0],
[0, 0, 0]],
[[1, 1, 1],
[1, 1, 1],
[1, 1, 1]],
[[0, 0, 0],
[0, 0, 0],
[0, 0, 0]]]
)
else:
img[:segmented_img.shape[0], :segmented_img.shape[1]] = segmented_img
struc = None
areas = label_image(img, structure=struc)
# A = np.zeros((np.prod(img.shape), areas[1]), dtype=bool)
selem: np.array = np.ones((selem,)*len(segmented_img.shape))
if areas[1] < 50:
print("Less than 50 regions, not parallelizing")
sparses = []
for i in tqdm(range(areas[1])):
sparses.append(
_get_sparse_mask(areas, i, edge_method, selem)
)
else:
print("Greater than 50 regions, parallelizing with joblib")
sparses = Parallel(n_jobs=cpu_count(), verbose=5)(
delayed(_get_sparse_mask)(areas, i, edge_method, selem) for i in range(areas[1])
)
if transpose:
return scipy.sparse.vstack(sparses).T.toarray()
else:
return scipy.sparse.vstack(sparses).toarray()
def get_closed_pores(bin_3d_img, structure_neighbors_num=6):
"""
function returns a bin_3d_img of closed pores for the
chosen neighbor voxels configuration
"""
if not np.asarray(bin_3d_img).dtype == bool:
bin_3d_img = bin_3d_img.astype(bool)
structure = get_structure(structure_neighbors_num)
connected_components, _ = label_image(bin_3d_img, structure)
levitating_volume = clear_border(connected_components) > 0
return levitating_volume
def filter_pores_mask(pore_mask_img, lowest_value, highest_value=None):
labeled_img, _ = label_image(pore_mask_img)
unique_labels, unique_counts = np.unique(labeled_img, return_counts=True)
if highest_value == None:
highest_value = np.max(unique_counts) + 1
accepted_labels = unique_labels[np.logical_and(
unique_labels > 0, unique_counts < highest_value,
unique_counts > lowest_value)]
mask = np.zeros(pore_mask_img.shape)
for elem in accepted_labels:
# TODO: replace with _create_mask_layer_for_label
mask += np.where(elem == labeled_img, True, False)
return mask
