def score_zernike(zf, radii, labels, indexes=None):
"""Score the output of construct_zernike_polynomials
zf - the output of construct_zernike_polynomials which is I x J x K
where K is the number of zernike polynomials computed
radii - a vector of the radius of each of N labeled objects
labels - a label matrix
outputs a N x K matrix of the scores of each of the Zernikes for
each labeled object.
"""
if indexes is None:
indexes = np.arange(1, np.max(labels) + 1, dtype=np.int32)
else:
indexes = np.array(indexes, dtype=np.int32)
radii = np.array(radii)
k = zf.shape[2]
n = np.product(radii.shape)
score = np.zeros((n, k))
if n == 0:
return score
areas = radii**2 * np.pi
for ki in range(k):
zfk = zf[:, :, ki]
real_score = scipy.ndimage.sum(zfk.real, labels, indexes)
real_score = fixup_scipy_ndimage_result(real_score)
imag_score = scipy.ndimage.sum(zfk.imag, labels, indexes)
imag_score = fixup_scipy_ndimage_result(imag_score)
one_score = np.sqrt(real_score**2 + imag_score**2) / areas
score[:, ki] = one_score
return score
def score_zernike(zf, radii, labels, indexes=None):
"""Score the output of construct_zernike_polynomials
zf - the output of construct_zernike_polynomials which is I x J x K
where K is the number of zernike polynomials computed
radii - a vector of the radius of each of N labeled objects
labels - a label matrix
outputs a N x K matrix of the scores of each of the Zernikes for
each labeled object.
"""
if indexes is None:
indexes = np.arange(1, np.max(labels) + 1, dtype=np.int32)
else:
indexes = np.array(indexes, dtype=np.int32)
radii = np.array(radii)
k = zf.shape[2]
n = np.product(radii.shape)
score = np.zeros((n, k))
if n == 0:
return score
areas = radii ** 2 * np.pi
for ki in range(k):
zfk = zf[:, :, ki]
real_score = scipy.ndimage.sum(zfk.real, labels, indexes)
real_score = fixup_scipy_ndimage_result(real_score)
imag_score = scipy.ndimage.sum(zfk.imag, labels, indexes)
imag_score = fixup_scipy_ndimage_result(imag_score)
one_score = np.sqrt(real_score ** 2 + imag_score ** 2) / areas
score[:, ki] = one_score
return score
def filter_using_image(self, workspace, mask):
'''Filter out connections using local intensity minima between objects
workspace - the workspace for the image set
mask - mask of background points within the minimum distance
'''
#
# NOTE: This is an efficient implementation and an improvement
# in accuracy over the Matlab version. It would be faster and
# more accurate to eliminate the line-connecting and instead
# do the following:
# * Distance transform to get the coordinates of the closest
# point in an object for points in the background that are
# at most 1/2 of the max distance between objects.
# * Take the intensity at this closest point and similarly
# label the background point if the background intensity
# is at least the minimum intensity fraction
# * Assume there is a connection between objects if, after this
# labeling, there are adjacent points in each object.
#
# As it is, the algorithm duplicates the Matlab version but suffers
# for cells whose intensity isn't high in the centroid and clearly
# suffers when two cells touch at some point that's off of the line
# between the two.
#
objects = workspace.object_set.get_objects(self.objects_name.value)
labels = objects.segmented
image = self.get_image(workspace)
if self.show_window:
# Save the image for display
workspace.display_data.image = image
#
# Do a distance transform into the background to label points
# in the background with their closest foreground object
#
i, j = scind.distance_transform_edt(labels == 0,
return_indices=True,
return_distances=False)
confluent_labels = labels[i, j]
confluent_labels[~mask] = 0
if self.where_algorithm == CA_CLOSEST_POINT:
#
# For the closest point method, find the intensity at
# the closest point in the object (which will be the point itself
# for points in the object).
#
object_intensity = image[i,
j] * self.minimum_intensity_fraction.value
confluent_labels[object_intensity > image] = 0
count, index, c_j = morph.find_neighbors(confluent_labels)
if len(c_j) == 0:
# Nobody touches - return the labels matrix
return labels
#
# Make a row of i matching the touching j
#
c_i = np.zeros(len(c_j))
#
# Eliminate labels without matches
#
label_numbers = np.arange(1, len(count) + 1)[count > 0]
index = index[count > 0]
count = count[count > 0]
#
# Get the differences between labels so we can use a cumsum trick
# to increment to the next label when they change
#
label_numbers[1:] = label_numbers[1:] - label_numbers[:-1]
c_i[index] = label_numbers
c_i = np.cumsum(c_i).astype(int)
if self.where_algorithm == CA_CENTROIDS:
#
# Only connect points > minimum intensity fraction
#
center_i, center_j = morph.centers_of_labels(labels)
indexes, counts, i, j = morph.get_line_pts(center_i[c_i - 1],
center_j[c_i - 1],
center_i[c_j - 1],
center_j[c_j - 1])
#
# The indexes of the centroids at pt1
#
last_indexes = indexes + counts - 1
#
# The minimum of the intensities at pt0 and pt1
#
centroid_intensities = np.minimum(
image[i[indexes], j[indexes]], image[i[last_indexes],
j[last_indexes]])
#
# Assign label numbers to each point so we can use
# scipy.ndimage.minimum. The label numbers are indexes into
# "connections" above.
#
pt_labels = np.zeros(len(i), int)
pt_labels[indexes[1:]] = 1
pt_labels = np.cumsum(pt_labels)
minima = scind.minimum(image[i, j], pt_labels,
np.arange(len(indexes)))
minima = morph.fixup_scipy_ndimage_result(minima)
#
# Filter the connections using the image
#
mif = self.minimum_intensity_fraction.value
i = c_i[centroid_intensities * mif <= minima]
j = c_j[centroid_intensities * mif <= minima]
else:
i = c_i
j = c_j
#
# Add in connections from self to self
#
unique_labels = np.unique(labels)
i = np.hstack((i, unique_labels))
j = np.hstack((j, unique_labels))
#
# Run "all_connected_components" to get a component # for
# objects identified as same.
#
new_indexes = morph.all_connected_components(i, j)
new_labels = np.zeros(labels.shape, int)
new_labels[labels != 0] = new_indexes[labels[labels != 0]]
return new_labels
def run(self, workspace):
"""Run the algorithm on one image set"""
#
# Get the image as a binary image
#
image_set = workspace.image_set
image = image_set.get_image(self.image_name.value, must_be_binary=True)
mask = image.pixel_data
if image.has_mask:
mask = mask & image.mask
angle_count = self.angle_count.value
#
# We collect the i,j and angle of pairs of points that
# are 3-d adjacent after erosion.
#
# i - the i coordinate of each point found after erosion
# j - the j coordinate of each point found after erosion
# a - the angle of the structuring element for each point found
#
i = numpy.zeros(0, int)
j = numpy.zeros(0, int)
a = numpy.zeros(0, int)
ig, jg = numpy.mgrid[0 : mask.shape[0], 0 : mask.shape[1]]
this_idx = 0
for angle_number in range(angle_count):
angle = float(angle_number) * numpy.pi / float(angle_count)
strel = self.get_diamond(angle)
erosion = binary_erosion(mask, strel)
#
# Accumulate the count, i, j and angle for all foreground points
# in the erosion
#
this_count = numpy.sum(erosion)
i = numpy.hstack((i, ig[erosion]))
j = numpy.hstack((j, jg[erosion]))
a = numpy.hstack((a, numpy.ones(this_count, float) * angle))
#
# Find connections based on distances, not adjacency
#
first, second = self.find_adjacent_by_distance(i, j, a)
#
# Do all connected components.
#
if len(first) > 0:
ij_labels = all_connected_components(first, second) + 1
nlabels = numpy.max(ij_labels)
label_indexes = numpy.arange(1, nlabels + 1)
#
# Compute the measurements
#
center_x = fixup_scipy_ndimage_result(
mean_of_labels(j, ij_labels, label_indexes)
)
center_y = fixup_scipy_ndimage_result(
mean_of_labels(i, ij_labels, label_indexes)
)
#
# The angles are wierdly complicated because of the wrap-around.
# You can imagine some horrible cases, like a circular patch of
# "worm" in which all angles are represented or a gentle "U"
# curve.
#
# For now, I'm going to use the following heuristic:
#
# Compute two different "angles". The angles of one go
# from 0 to 180 and the angles of the other go from -90 to 90.
# Take the variance of these from the mean and
# choose the representation with the lowest variance.
#
# An alternative would be to compute the variance at each possible
# dividing point. Another alternative would be to actually trace through
# the connected components - both overkill for such an inconsequential
# measurement I hope.
#
angles = fixup_scipy_ndimage_result(
mean_of_labels(a, ij_labels, label_indexes)
)
vangles = fixup_scipy_ndimage_result(
mean_of_labels(
(a - angles[ij_labels - 1]) ** 2, ij_labels, label_indexes
)
)
aa = a.copy()
aa[a > numpy.pi / 2] -= numpy.pi
aangles = fixup_scipy_ndimage_result(
mean_of_labels(aa, ij_labels, label_indexes)
)
vaangles = fixup_scipy_ndimage_result(
mean_of_labels(
(aa - aangles[ij_labels - 1]) ** 2, ij_labels, label_indexes
)
)
aangles[aangles < 0] += numpy.pi
angles[vaangles < vangles] = aangles[vaangles < vangles]
#
# Squish the labels to 2-d. The labels for overlaps are arbitrary.
#
labels = numpy.zeros(mask.shape, int)
labels[i, j] = ij_labels
else:
center_x = numpy.zeros(0, int)
center_y = numpy.zeros(0, int)
angles = numpy.zeros(0)
nlabels = 0
label_indexes = numpy.zeros(0, int)
labels = numpy.zeros(mask.shape, int)
m = workspace.measurements
assert isinstance(m, Measurements)
object_name = self.object_name.value
m.add_measurement(object_name, M_LOCATION_CENTER_X, center_x)
m.add_measurement(object_name, M_LOCATION_CENTER_Y, center_y)
m.add_measurement(object_name, M_ANGLE, angles * 180 / numpy.pi)
m.add_measurement(
object_name, M_NUMBER_OBJECT_NUMBER, label_indexes,
)
m.add_image_measurement(FF_COUNT % object_name, nlabels)
#
# Make the objects
#
object_set = workspace.object_set
assert isinstance(object_set, ObjectSet)
objects = Objects()
objects.segmented = labels
objects.parent_image = image
object_set.add_objects(objects, object_name)
if self.show_window:
workspace.display_data.i = center_y
workspace.display_data.j = center_x
workspace.display_data.angle = angles
workspace.display_data.mask = mask
workspace.display_data.labels = labels
workspace.display_data.count = nlabels
def filter_using_image(self, workspace, mask):
'''Filter out connections using local intensity minima between objects
workspace - the workspace for the image set
mask - mask of background points within the minimum distance
'''
#
# NOTE: This is an efficient implementation and an improvement
# in accuracy over the Matlab version. It would be faster and
# more accurate to eliminate the line-connecting and instead
# do the following:
# * Distance transform to get the coordinates of the closest
# point in an object for points in the background that are
# at most 1/2 of the max distance between objects.
# * Take the intensity at this closest point and similarly
# label the background point if the background intensity
# is at least the minimum intensity fraction
# * Assume there is a connection between objects if, after this
# labeling, there are adjacent points in each object.
#
# As it is, the algorithm duplicates the Matlab version but suffers
# for cells whose intensity isn't high in the centroid and clearly
# suffers when two cells touch at some point that's off of the line
# between the two.
#
objects = workspace.object_set.get_objects(self.objects_name.value)
labels = objects.segmented
image = self.get_image(workspace)
if self.show_window:
# Save the image for display
workspace.display_data.image = image
#
# Do a distance transform into the background to label points
# in the background with their closest foreground object
#
i, j = scind.distance_transform_edt(labels==0,
return_indices=True,
return_distances=False)
confluent_labels = labels[i,j]
confluent_labels[~mask] = 0
if self.where_algorithm == CA_CLOSEST_POINT:
#
# For the closest point method, find the intensity at
# the closest point in the object (which will be the point itself
# for points in the object).
#
object_intensity = image[i,j] * self.minimum_intensity_fraction.value
confluent_labels[object_intensity > image] = 0
count, index, c_j = morph.find_neighbors(confluent_labels)
if len(c_j) == 0:
# Nobody touches - return the labels matrix
return labels
#
# Make a row of i matching the touching j
#
c_i = np.zeros(len(c_j))
#
# Eliminate labels without matches
#
label_numbers = np.arange(1,len(count)+1)[count > 0]
index = index[count > 0]
count = count[count > 0]
#
# Get the differences between labels so we can use a cumsum trick
# to increment to the next label when they change
#
label_numbers[1:] = label_numbers[1:] - label_numbers[:-1]
c_i[index] = label_numbers
c_i = np.cumsum(c_i).astype(int)
if self.where_algorithm == CA_CENTROIDS:
#
# Only connect points > minimum intensity fraction
#
center_i, center_j = morph.centers_of_labels(labels)
indexes, counts, i, j = morph.get_line_pts(
center_i[c_i-1], center_j[c_i-1],
center_i[c_j-1], center_j[c_j-1])
#
# The indexes of the centroids at pt1
#
last_indexes = indexes+counts-1
#
# The minimum of the intensities at pt0 and pt1
#
centroid_intensities = np.minimum(
image[i[indexes],j[indexes]],
image[i[last_indexes], j[last_indexes]])
#
# Assign label numbers to each point so we can use
# scipy.ndimage.minimum. The label numbers are indexes into
# "connections" above.
#
pt_labels = np.zeros(len(i), int)
pt_labels[indexes[1:]] = 1
pt_labels = np.cumsum(pt_labels)
minima = scind.minimum(image[i,j], pt_labels, np.arange(len(indexes)))
minima = morph.fixup_scipy_ndimage_result(minima)
#
# Filter the connections using the image
#
mif = self.minimum_intensity_fraction.value
i = c_i[centroid_intensities * mif <= minima]
j = c_j[centroid_intensities * mif <= minima]
else:
i = c_i
j = c_j
#
# Add in connections from self to self
#
unique_labels = np.unique(labels)
i = np.hstack((i, unique_labels))
j = np.hstack((j, unique_labels))
#
# Run "all_connected_components" to get a component # for
# objects identified as same.
#
new_indexes = morph.all_connected_components(i, j)
new_labels = np.zeros(labels.shape, int)
new_labels[labels != 0] = new_indexes[labels[labels != 0]]
return new_labels
def run(self, workspace):
"""Run the module on an image set"""
object_name = self.object_name.value
remaining_object_name = self.remaining_objects.value
original_objects = workspace.object_set.get_objects(object_name)
if self.mask_choice == MC_IMAGE:
mask = workspace.image_set.get_image(
self.masking_image.value, must_be_binary=True
)
mask = mask.pixel_data
else:
masking_objects = workspace.object_set.get_objects(
self.masking_objects.value
)
mask = masking_objects.segmented > 0
if self.wants_inverted_mask:
mask = ~mask
#
# Load the labels
#
labels = original_objects.segmented.copy()
nobjects = numpy.max(labels)
#
# Resize the mask to cover the objects
#
mask, m1 = size_similarly(labels, mask)
mask[~m1] = False
#
# Apply the mask according to the overlap choice.
#
if nobjects == 0:
pass
elif self.overlap_choice == P_MASK:
labels = labels * mask
else:
pixel_counts = fixup_scipy_ndimage_result(
scipy.ndimage.sum(
mask, labels, numpy.arange(1, nobjects + 1, dtype=numpy.int32)
)
)
if self.overlap_choice == P_KEEP:
keep = pixel_counts > 0
else:
total_pixels = fixup_scipy_ndimage_result(
scipy.ndimage.sum(
numpy.ones(labels.shape),
labels,
numpy.arange(1, nobjects + 1, dtype=numpy.int32),
)
)
if self.overlap_choice == P_REMOVE:
keep = pixel_counts == total_pixels
elif self.overlap_choice == P_REMOVE_PERCENTAGE:
fraction = self.overlap_fraction.value
keep = pixel_counts / total_pixels >= fraction
else:
raise NotImplementedError(
"Unknown overlap-handling choice: %s", self.overlap_choice.value
)
keep = numpy.hstack(([False], keep))
labels[~keep[labels]] = 0
#
# Renumber the labels matrix if requested
#
if self.retain_or_renumber == R_RENUMBER:
unique_labels = numpy.unique(labels[labels != 0])
indexer = numpy.zeros(nobjects + 1, int)
indexer[unique_labels] = numpy.arange(1, len(unique_labels) + 1)
labels = indexer[labels]
parent_objects = unique_labels
else:
parent_objects = numpy.arange(1, nobjects + 1)
#
# Add the objects
#
remaining_objects = Objects()
remaining_objects.segmented = labels
remaining_objects.unedited_segmented = original_objects.unedited_segmented
workspace.object_set.add_objects(remaining_objects, remaining_object_name)
#
# Add measurements
#
m = workspace.measurements
m.add_measurement(
remaining_object_name, FF_PARENT % object_name, parent_objects,
)
if numpy.max(original_objects.segmented) == 0:
child_count = numpy.array([], int)
else:
child_count = fixup_scipy_ndimage_result(
scipy.ndimage.sum(
labels,
original_objects.segmented,
numpy.arange(1, nobjects + 1, dtype=numpy.int32),
)
)
child_count = (child_count > 0).astype(int)
m.add_measurement(
object_name, FF_CHILDREN_COUNT % remaining_object_name, child_count,
)
if self.retain_or_renumber == R_RETAIN:
remaining_object_count = nobjects
else:
remaining_object_count = len(unique_labels)
add_object_count_measurements(m, remaining_object_name, remaining_object_count)
add_object_location_measurements(m, remaining_object_name, labels)
#
# Save the input, mask and output images for display
#
if self.show_window:
workspace.display_data.original_labels = original_objects.segmented
workspace.display_data.final_labels = labels
workspace.display_data.mask = mask
