def watershed(self, Ta=0):
"""
Identification of particles through inverted slope comparisons
Parameters
-----------
Ta : int
Threshold value in which the particles will be identified by
"""
self.Ta = Ta
dist = mh.distance(self.image > 0.05 * self.Ta)
dist1 = dist
dist = dist.max() - dist
dist -= dist.min() # inverting color
dist = dist / float(dist.ptp()) * 255
dist = dist.astype(np.uint8)
self.dist = mh.stretch(dist, 0, 255)
self.labels, self.n_particles = mh.label(self.image > 0.7 * self.Ta)
thr = np.median(dist)
# not accurate to particles detected(?)
# but matches dist graph well
thresh = (dist < thr)
areas = 0.9 * mh.cwatershed(dist, self.labels)
self.areas = areas * thresh
return
def buildGraph(img, skel):
print('Building tree...')
G = nx.Graph()
visited = np.zeros_like(
skel) # a new array of zeros, same dimensions as skel
print(' distance transform...')
dmap = mh.distance(img)
print(' constructing tree...')
buildTree(skel, visited, dmap, root, j, t, G)
# measure node diameters
measureDia(G, dmap)
# automatically remove bad nodes and branches, e.g. too small ones
removed = cleanup(G, root)
# show (automatically) removed nodes
for x, y in removed:
plt.gca().add_patch(plt.Circle((x, y), radius=4, alpha=.4))
updateMeasures(G, root)
# measure node diameters 2
measureLeafWidth(contour, G, root)
initializeColorNode(G)
print('Done.')
return G
def extract(self):
'''Extracts point pattern features.
Returns
-------
pandas.DataFrame
extracted feature values for each object in
:attr:`label_image <jtlib.features.PointPattern.label_image>`
'''
logger.info('extract point pattern features')
features = dict()
for obj in self.parent_object_ids:
parent_obj_img = self.get_parent_object_mask_image(obj)
points_img = self.get_points_object_label_image(obj)
point_ids = np.unique(points_img)[1:]
mh.labeled.relabel(points_img, inplace=True)
size = np.sum(parent_obj_img)
abs_border_dist_img = mh.distance(parent_obj_img).astype(float)
rel_border_dist_img = abs_border_dist_img / size
centroids = mh.center_of_mass(points_img, labels=points_img)
centroids = centroids[1:, :].astype(int)
abs_distance_matrix = squareform(pdist(centroids))
rel_distance_matrix = abs_distance_matrix / size
indexer = np.arange(centroids.shape[0])
if len(indexer) == 0:
continue
if len(indexer) == 1:
y, x = centroids[0, :]
values = [
abs_border_dist_img[y, x], rel_border_dist_img[y, x],
np.nan, np.nan, np.nan, np.nan, np.nan, np.nan
]
features[point_ids[0]] = values
continue
for i, (y, x) in enumerate(centroids):
idx = indexer != i
values = [
abs_border_dist_img[y, x],
rel_border_dist_img[y, x],
np.nanmin(abs_distance_matrix[i, idx]),
np.nanmin(rel_distance_matrix[i, idx]),
np.nanmean(abs_distance_matrix[i, idx]),
np.nanstd(abs_distance_matrix[i, idx]),
np.nanmean(rel_distance_matrix[i, idx]),
np.nanstd(rel_distance_matrix[i, idx]),
]
features[point_ids[i]] = values
ids = features.keys()
values = list()
nans = [np.nan for _ in range(len(self.names))]
for i in self.object_ids:
if i not in ids:
logger.warn('values missing for object #%d', i)
features[i] = nans
values.append(features[i])
return pd.DataFrame(values, columns=self.names, index=self.object_ids)
def get_seeds(boundary, method='grid', next_id=1):
if method == 'grid':
height = boundary.shape[0]
width = boundary.shape[1]
seed_positions = np.ogrid[0:height:seed_distance,
0:width:seed_distance]
num_seeds_y = seed_positions[0].size
num_seeds_x = seed_positions[1].size
num_seeds = num_seeds_x * num_seeds_y
seeds = np.zeros_like(boundary).astype(np.int32)
seeds[seed_positions] = np.arange(next_id,
next_id + num_seeds).reshape(
(num_seeds_y, num_seeds_x))
if method == 'minima':
minima = mahotas.regmin(boundary)
seeds, num_seeds = mahotas.label(minima)
seeds += next_id
seeds[seeds == next_id] = 0
if method == 'maxima_distance':
distance = mahotas.distance(boundary < 0.5)
maxima = mahotas.regmax(distance)
seeds, num_seeds = mahotas.label(maxima)
seeds += next_id
seeds[seeds == next_id] = 0
return seeds, num_seeds
def chromatids_elements(TopHatedChromosome):
'''Take a High pass filtered (or top hat) image of a chromosome and label the chromatids elements
'''
threshed = TopHatedChromosome > 0
#threshed = mh.open(threshed)
labthres, _ = mh.label(threshed)
labsz = mh.labeled.labeled_size(labthres)
mh.labeled.remove_regions_where(labthres, labsz < 2, inplace=True)
threshed = labthres > 0
skel2 = mh.thin(threshed)
bp2 = branchedPoints(skel2, showSE=False) > 0
rem = np.logical_and(skel2, np.logical_not(bp2))
labskel, _ = mh.labeled.label(rem)
#print labskel.dtype
size_sk = mh.labeled.labeled_size(labskel)
#print size_sk
skelem = mh.labeled.remove_regions_where(labskel, size_sk < 4)
distances = mh.stretch(mh.distance(threshed))
surface = (distances.max() - distances)
chr_label = mh.cwatershed(surface, skelem)
#print chr_label.dtype, type(chr_label)
chr_label *= threshed
#This convertion is important !!
chr_label = chr_label.astype(np.intc)
#-------------------------------
mh.labeled.relabel(chr_label, inplace=True)
labsize2 = mh.labeled.labeled_size(chr_label)
cleaned = mh.labeled.remove_regions_where(chr_label, labsize2 < 8)
mh.labeled.relabel(cleaned, inplace=True)
return cleaned
def perform_watershed(threshed, maxima):
distances = mh.stretch(mh.distance(threshed))
spots, n_spots = mh.label(maxima, Bc=np.ones((3, 3)))
surface = (distances.max() - distances)
return sk.morphology.watershed(surface, spots, mask=threshed)
def test_4d():
np.random.seed(324)
for _ in range(16):
binim = np.random.random((4,8,4,6)) > .5
dist = distance(binim)
assert dist.shape == binim.shape
assert np.all(dist[~binim] == 0)
assert np.all(dist == _slow_dist4d(binim, 'euclidean2'))
def test_4d():
np.random.seed(324)
for _ in range(16):
binim = np.random.random((4, 8, 4, 6)) > .5
dist = distance(binim)
assert dist.shape == binim.shape
assert np.all(dist[~binim] == 0)
assert np.all(dist == _slow_dist4d(binim, 'euclidean2'))
def segment(img, T):
binimg = (img > T)
binimg = ndimage.median_filter(binimg, size=5)
dist = mahotas.distance(binimg)
dist = dist.astype(np.int32)
maxima = pymorph.regmax(dist)
maxima,_ = ndimage.label(maxima)
return mahotas.cwatershed(dist.max() - dist, maxima)
def _distance_transform_seeds(pmap, threshold, start_id):
distance = mahotas.distance(pmap < threshold)
maxima = mahotas.regmax(distance)
seeds, num_seeds = mahotas.label(maxima)
seeds += start_id
#seeds[seeds==start_id] = 0 # TODO I don't get this
return seeds, num_seeds
def segmentCones(self):
"""Function to take regional maxima and segment the cones from them"""
dist = mh.distance(self.orgImage > self.params['threshold'])
dist = dist.max() - dist
dist = dist = dist - dist.min()
dist = dist/float(dist.ptp()) * 255
dist = dist.astype(np.uint8)
self.params['cones'] = mh.cwatershed(dist, self.ConeCounts.Seeds)
def skel(self, img):
img[0,:]=0 # make 1st line in the image black to achieve consistent result between distance field and medial axis skeleton.
img[len(img)-1,:]=0 # make last line in the image black to achieve consistent result between distance field and medial axis skeleton.
img[:,len(img[0])-1]=0 # make right column in the image black to achieve consistent result between distance field and medial axis skeleton.
img[:,0]=0 # make left column in the image black to achieve consistent result between distance field and medial axis skeleton.
dmap = m.distance(img>0,metric='euclidean')
dmap=np.sqrt(dmap)*2
skelImg=m.thin(img>0)
return skelImg, dmap
def nuclei_regions(comp_map):
"""
NUCLEI_REGIONS: extract "support regions" for nuclei. This function
expects as input a "tissue components map" (as returned, for example,
by segm.tissue_components) where values of 1 indicate pixels having
a color corresponding to nuclei.
It returns a set of compact support regions corresponding to the
nuclei.
:param comp_map: numpy.ndarray
A mask identifying different tissue components, as obtained
by classification in RGB space. The value 0
See segm.tissue.tissue_components()
:return:
"""
# Deprecated:...
# img_hem, _ = rgb2he(img0, normalize=True)
# img_hem = denoise_tv_bregman(img_hem, HE_OPTS['bregm'])
# Get a mask of nuclei regions by unsupervised clustering:
# Vector Quantization: background, mid-intensity Hem and high intensity Hem
# -train the quantizer for 3 levels
# vq = KMeans(n_clusters=3)
# vq.fit(img_hem.reshape((-1,1)))
# -the level of interest is the brightest:
# k = np.argsort(vq.cluster_centers_.squeeze())[2]
# mask_hem = (vq.labels_ == k).reshape(img_hem.shape)
# ...end deprecated
# Final mask:
mask = (comp_map == 1) # use the components classified by color
# mask = morph.closing(mask, selem=HE_OPTS['strel1'])
# mask = morph.opening(mask, selem=HE_OPTS['strel1'])
# morph.remove_small_objects(mask, in_place=True)
# mask = (mask > 0)
mask = mahotas.close_holes(mask)
morph.remove_small_objects(mask, in_place=True)
dst = mahotas.stretch(mahotas.distance(mask))
Bc=np.ones((9,9))
lmax = mahotas.regmax(dst, Bc=Bc)
spots, _ = mahotas.label(lmax, Bc=Bc)
regions = mahotas.cwatershed(lmax.max() - lmax, spots) * mask
return regions
# end NUCLEI_REGIONS
def watershed_single_channel(np_array, threshold=10, blur_factor=9, max_bb_size=13000, min_bb_size=1000, footprint=10):
# INPUT:
# np_array: numpy image of single channel
# threshold: minimum value to be analysed
# blur_factor: gaussian blur. some blur is good. too much is bad. too little is bad too.
# max_bb_size: set maximum size for a bounding box.
# min_bb_size: set minimum size for a bounding box.
# footprint: box of size footprint x footprint used to find regions of maximum intensity.
# OUTPUT:
# return: array of bounding boxes for the channel image given
# NOTE: input values should be tuned. this alg is useless if input arguments are not optimized... right now its by hand. otherwise just leave them
# nuclear is a blured version of origional image
nuclear = mh.gaussian_filter(np_array, blur_factor)
# calculate a minimum threshold using otsu method
otsu_thresh = threshold_otsu(nuclear)
# determine minimum threshold from otsu and input argument. if little/no signal in image; otsu value can be way too low
set_thresh = None
if threshold>otsu_thresh:
set_thresh = threshold
else:
set_thresh = otsu_thresh
# set values lower than set_thresh to zero
index_otsu = nuclear < set_thresh
nuclear[index_otsu] = 0
# determine areas of maximum intensity and the distance between them
thresh = (nuclear > nuclear.mean())
dist = mh.stretch(mh.distance(thresh))
Bc = np.ones((footprint, footprint))
# the code the generate region_props from watersheding alg.
maxima = mh.morph.regmax(dist, Bc=Bc)
spots, n_spots = mh.label(maxima, Bc=Bc)
sizes = mh.labeled.labeled_size(spots)
too_big = np.where(sizes > max_bb_size)
spots = mh.labeled.remove_regions(spots, too_big)
spots = mh.labeled.remove_bordering(spots)
spots, n_left = mh.labeled.relabel(spots)
surface = (dist.max() - dist)
areas = mh.cwatershed(surface, spots)
areas *= thresh
# get list of region properties from watershed. allot of information in region_props. allot of which is inaccurate. NEVER TRUST REGIONPROPS!
region_props=regionprops(areas,intensity_image=nuclear)
# generate array of bounding boxes from measured region properties. call bbs_from_rprops()
watershed_bb_array = bbs_from_rprops(region_props, max_bb_size, min_bb_size)
return(watershed_bb_array)
def watershed(Image_Current_T, CellDiam):
#If there are pixels above the threshold proceed to watershed
if Image_Current_T.max() == True:
#Create distance transform from thresholded image to help identify cell seeds
Image_Current_Tdist = mh.distance(Image_Current_T)
Image_Current_Tdist[Image_Current_Tdist < CellDiam * 0.3] = 0
#Define Sure Background for watershed
#Background is dilated proportional to cell diam. Allows final cell sizes to be a bit larger at end.
#Will not affect cell number but can influence overlap
#See https://docs.opencv.org/3.4/d3/db4/tutorial_py_watershed.html for tutorial that helps explain this
Dilate_Iterations = int(CellDiam // 2)
Dilate_bc = np.ones(
(3, 3)) #Use square structuring element instead of cross
Image_Current_SureBackground = Image_Current_T
for j in range(Dilate_Iterations):
Image_Current_SureBackground = mh.dilate(
Image_Current_SureBackground, Bc=Dilate_bc)
#Create seeds/foreground for watershed
#See https://docs.opencv.org/3.4/d3/db4/tutorial_py_watershed.html for tutorial that helps explain this
Regmax_bc = np.ones((CellDiam, CellDiam))
Image_Current_Seeds = mh.locmax(Image_Current_Tdist, Bc=Regmax_bc)
Image_Current_Seeds[Image_Current_Tdist == 0] = False
Image_Current_Seeds = mh.dilate(Image_Current_Seeds, np.ones((3, 3)))
seeds, nr_nuclei = mh.label(Image_Current_Seeds, Bc=np.ones((3, 3)))
Image_Current_Unknown = Image_Current_SureBackground.astype(
int) - Image_Current_Seeds.astype(int)
seeds += 1
seeds[Image_Current_Unknown == 1] = 0
#Perform watershed
Image_Current_Watershed = mh.cwatershed(
surface=Image_Current_SureBackground, markers=seeds)
Image_Current_Watershed -= 1 #Done so that background value is equal to 0.
Image_Current_Cells = Image_Current_Watershed
#If there are no pixels above the threshold watershed procedure has issues. Set cell count to 0.
elif Image_Current_T.max() == False:
Image_Current_Cells = Image_Current_T.astype(int)
nr_nuclei = 0
#return Image_Current_Seeds, nr_nuclei
return Image_Current_Cells, nr_nuclei
def _segment(cell):
# takes a numpy array of a microscopy
# segments it based on filtering the image then applying a distance transform and
# a watershed method to get the proper segmentation
import mahotas as mh
filt_cell = mh.gaussian_filter(cell, 2)
T = mh.thresholding.otsu((np.rint(filt_cell).astype('uint8')))
dist = mh.stretch(mh.distance(filt_cell > T))
Bc = np.ones((3,3))
rmax = mh.regmin((dist))
rmax = np.invert(rmax)
labels, num_cells = mh.label(rmax, Bc)
surface = (dist.max() - dist)
areas = mh.cwatershed(dist, labels)
areas *= T
return areas
def _segment(cell):
# takes a numpy array of a microscopy
# segments it based on filtering the image then applying a distance transform and
# a watershed method to get the proper segmentation
import mahotas as mh
filt_cell = mh.gaussian_filter(cell, 2)
T = mh.thresholding.otsu((np.rint(filt_cell).astype('uint8')))
dist = mh.stretch(mh.distance(filt_cell > T))
Bc = np.ones((3, 3))
rmax = mh.regmin((dist))
rmax = np.invert(rmax)
labels, num_cells = mh.label(rmax, Bc)
surface = (dist.max() - dist)
areas = mh.cwatershed(dist, labels)
areas *= T
return areas
def run(self):
while not self.terminate:
idx, x, y, value = self.queue.get()
self.zero[int(y), int(x)] = False
dmap = mh.distance(self.zero)
bools = dmap < self.brush_size_sq
#self.binary_array[idx][bools] = value
np.putmask(self.binary_array[idx], bools, value)
self.edited[idx] = True
#self.iw.overlay_item.setImage(self.binary_array[idx])
self.iw.olay_updated.emit(self.binary_array[idx])
self.zero[int(y), int(x)] = True
self.queue.task_done()
self.terminate = False
return 0
def segment(fname):
dna = mh.imread(fname)
dna = dna[:,:,0]
sigma = 12.
dnaf = mh.gaussian_filter(dna, sigma)
T_mean = dnaf.mean()
bin_image = dnaf > T_mean
labeled, nr_objects = mh.label(bin_image)
maxima = mh.regmax(mh.stretch(dnaf))
maxima = mh.dilate(maxima, np.ones((5,5)))
maxima,_ = mh.label(maxima)
dist = mh.distance(bin_image)
dist = 255 - mh.stretch(dist)
watershed = mh.cwatershed(dist, maxima)
watershed *= bin_image
return watershed
def identify_primary_objects(image, footprint=(16, 16), sigma=1):
"""
Identifies primary components in an image.
"""
response = skimage.filters.gaussian(image, sigma)
mask = response > skimage.filters.threshold_li(response)
image = mahotas.distance(mask)
image = skimage.exposure.rescale_intensity(image)
footprint = numpy.ones(footprint)
markers = mahotas.regmax(image, footprint)
markers, _ = mahotas.label(markers, footprint)
image = numpy.max(image) - image
return skimage.segmentation.watershed(image, markers, mask=mask)
def segment_nuc(im): T = mh.thresholding.otsu(im) # calculate a threshold value # Apply a gaussian filter to smooth the image smoothed = mh.gaussian_filter(im, 8) thresholded= smoothed > T # do threshold # Watershed smoothed = mh.gaussian_filter(im, 10) regional_max = mh.regmax(smoothed) dist_im = mh.distance(thresholded) seeds,count = mh.label(regional_max) # nuclei count watershedded = mh.cwatershed(dist_im, seeds) # Remove areas that aren't nuclei watershedded[np.logical_not(thresholded)] = 0 cell_id = ['abc', 'def', 'ghi'] # cell_centroid = [(1,2),(3,4)] # return watershedded, cell_id return watershedded
def buildGraph(img, skel):
print('Building tree...')
G = nx.Graph()
visited = np.zeros_like(skel) # a new array of zeros, same dimensions as skel
print(' distance transform...')
dmap = mh.distance(img)
print(' constructing tree...')
buildTree(skel, visited, dmap, root, j, t, G)
# measure node diameters
measureDia(G, dmap)
# automatically remove bad nodes and branches, e.g. too small ones
removed = cleanup(G, root)
# show (automatically) removed nodes
for x,y in removed:
plt.gca().add_patch(plt.Circle((x,y), radius=4, alpha=.4))
updateMeasures(G, root)
print('Done.')
return G
from __future__ import print_function import pylab as p import numpy as np import mahotas f = np.ones((256,256), bool) f[200:,240:] = False f[128:144,32:48] = False # f is basically True with the exception of two islands: one in the lower-right # corner, another, middle-left dmap = mahotas.distance(f) p.imshow(dmap) p.show()
#%%
desc = []
labels = []
for i in range(0, 3, 1):
im = skimage.io.imread('labeled/{0}.bmp'.format(i))
im = im[:(im.shape[0] / b) * b, :(im.shape[1] / b) * b]
plt.imshow(im)
lg = skimage.io.imread('labeled/{0}_map.png'.format(i))
lg = lg[:(im.shape[0] / b) * b, :(im.shape[1] / b) * b]
lg = lg[:, :, 3]
lg = mahotas.distance(lg) > 100
descriptors, (nhd, nwd), (dh, dw) = build_descriptors(im, b)
fig = plt.gcf()
for ii in range(nhd):
for jj in range(nwd):
desc.append(descriptors[ii * nwd + jj])
labels.append(lg[ii * b + dh / 2 + b / 2, jj * b + dw / 2 + b / 2])
if labels[-1]:
circle = plt.Circle((jj * b + dw / 2 + b / 2, ii * b + dh / 2 + b / 2), 1.0, color='r')
fig.gca().add_artist(circle)
plt.show()
svm = sklearn.svm.LinearSVC(class_weight = 'balanced')
def compare_slow(bw):
for metric in ('euclidean', 'euclidean2'):
f = distance(bw, metric)
sd = _slow_dist(bw, metric)
assert np.all(f == sd)
) # Funckija mh.gaussian_filter() postavlja vrijednost varijable u float64.
rmax = mh.regmax(result) # Trazi maksimalne vrijednosti.
print rmax
pylab.imshow(mh.overlay(image, rmax))
pylab.show()
labeled, nr_objects = mh.label(rmax)
# mh.overlay() funckija postavlja img kao pozadinu a preko nje postavlja vrijednosti variable rmax u crvenom kanalu.
print('Broj pronadenih objekata je {}.'.format(nr_objects))
pylab.imshow(labeled)
# pylab.gray()
pylab.show()
dist = mh.distance(result)
dist = dist.max() - dist
dist -= dist.min()
dist = dist / float(dist.ptp()) * 255
dist = dist.astype(np.uint8)
objects = mh.cwatershed(dist, labeled)
whole = mh.segmentation.gvoronoi(
objects
) # Voronoi segemtacija, svaki piksel poprima vrijednost najblizeg maksimuma.
pylab.imshow(objects)
pylab.show()
# print (labeled).shape
# print (labeled).max()
# print (labeled).min()
dnaf = mh.gaussian_filter(dnaf, 8)
rmax = mh.regmax(dnaf)
pylab.imshow(mh.overlay(
dna, rmax)) # print dna and rmax (with second channel in red)
pylab.show() # seeds only show when image is zoomed in
dnaf = mh.gaussian_filter(dnaf,
16) # apply different filter to yield better result
rmax = mh.regmax(dnaf)
pylab.imshow(mh.overlay(dna, rmax))
pylab.show()
seeds, nr_nuclei = mh.label(rmax) # nuclei count
print nr_nuclei # unlike the example, the result is 36 compared to 22
dist = mh.distance(dnat)
dist = dist.max() - dist
dist -= dist.min()
dist = dist / float(dist.ptp()) * 255
dist = dist.astype(np.uint8)
pylab.imshow(dist)
pylab.show()
nuclei = mh.cwatershed(dist, seeds)
pylab.imshow(nuclei)
pylab.show()
pylab.show()
nuclei[nuclei1] = 0
pylab.imshow(nuclei1)
pylab.show()
def process_image(im, d, test=False, remove_bordering=False):
plt.figure(1, frameon=False)
sigma = 75
blurred = mh.gaussian_filter(im.astype(float), sigma)
T_mean = blurred.mean()
bin_image = im > T_mean
maxima = mh.regmax(mh.stretch(blurred))
maxima, _ = mh.label(maxima)
dist = mh.distance(bin_image)
dist = 255 - mh.stretch(dist)
watershed = mh.cwatershed(dist, maxima)
_, old_nr_objects = mh.labeled.relabel(watershed)
sizes = mh.labeled.labeled_size(watershed)
min_size = 100000
filtered = mh.labeled.remove_regions_where(
watershed * bin_image, sizes < min_size)
_, nr_objects = mh.labeled.relabel(filtered)
print('Removed', old_nr_objects - nr_objects, 'small regions')
old_nr_objects = nr_objects
if (remove_bordering):
filtered = mh.labeled.remove_bordering(filtered)
labeled, nr_objects = mh.labeled.relabel(filtered)
print('Removed', old_nr_objects - nr_objects, 'bordering cells')
print("Number of cells: {}".format(nr_objects))
fin_weights = mh.labeled_sum(im.astype(np.uint32), labeled)
fin_sizes = mh.labeled.labeled_size(labeled)
fin_result = fin_weights / fin_sizes
if (test):
f, axarr = plt.subplots(2, 2)
for i in range(2):
for j in range(2):
axarr[i][j].axis('off')
axarr[0, 0].imshow(im)
axarr[0, 0].set_title('Source')
axarr[0, 1].imshow(labeled)
axarr[0, 1].set_title('Labeled')
axarr[1, 0].imshow(watershed)
axarr[1, 0].set_title('Watershed')
axarr[1, 1].imshow(blurred)
axarr[1, 1].set_title('Blurred')
for i in range(1, nr_objects + 1):
print("Cell {} average luminescence is {}".format(
i, fin_result[i]))
bbox = mh.bbox((labeled == i))
plt.text((bbox[2] + bbox[3]) / 2, (bbox[0] + bbox[1]
) / 2, str(i), fontsize=20, color='black')
# plt.show()
plt.savefig("test" + str(nr_objects) + ".svg",
format='svg', bbox_inches='tight', dpi=1200)
else:
for i in range(1, nr_objects + 1):
bbox = mh.bbox((labeled == i))
cell = (im * (labeled == i))[bbox[0]:bbox[1], bbox[2]:bbox[3]]
hashed = hashlib.sha1(im).hexdigest()
imsave(d + data_dir + hashed + '-' + str(i) +
'.png', imresize(cell, (img_rows, img_cols)))
def separate_clumped_objects(clumps_image, min_cut_area, min_area, max_area,
max_circularity, max_convexity):
'''Separates objects in `clumps_image` based on morphological criteria.
Parameters
----------
clumps_image: numpy.ndarray[Union[numpy.int32, numpy.bool]]
objects that should be separated
min_cut_area: int
minimal area an object must have (prevents cuts that would result
in too small objects)
min_area: int
minimal area an object must have to be considered a clump
max_area: int
maximal area an object can have to be considered a clump
max_circularity: float
maximal circularity an object must have to be considerd a clump
max_convexity: float
maximal convexity an object must have to be considerd a clump
Returns
-------
numpy.ndarray[numpy.uint32]
separated objects
See also
--------
:class:`jtlib.features.Morphology`
'''
logger.info('separate clumped objects')
label_image, n_objects = mh.label(clumps_image)
if n_objects == 0:
logger.debug('no objects')
return label_image
pad = 1
cutting_pass = 1
separated_image = label_image.copy()
while True:
logger.info('cutting pass #%d', cutting_pass)
cutting_pass += 1
label_image = mh.label(label_image > 0)[0]
f = Morphology(label_image)
values = f.extract()
index = (
(min_area < values['Morphology_Area']) &
(values['Morphology_Area'] <= max_area) &
(values['Morphology_Convexity'] <= max_convexity) &
(values['Morphology_Circularity'] <= max_circularity)
)
clumped_ids = values[index].index.values
not_clumped_ids = values[~index].index.values
if len(clumped_ids) == 0:
logger.debug('no more clumped objects')
break
mh.labeled.remove_regions(label_image, not_clumped_ids, inplace=True)
mh.labeled.relabel(label_image, inplace=True)
bboxes = mh.labeled.bbox(label_image)
for oid in np.unique(label_image[label_image > 0]):
bbox = bboxes[oid]
logger.debug('process clumped object #%d', oid)
obj_image = extract_bbox(label_image, bboxes[oid], pad=pad)
obj_image = obj_image == oid
# Rescale distance intensities to make them independent of clump size
dist = mh.stretch(mh.distance(obj_image))
# Find peaks that can be used as seeds for the watershed transform
thresh = mh.otsu(dist)
peaks = dist > thresh
n = mh.label(peaks)[1]
if n == 1:
logger.debug(
'only one peak detected - perform iterative erosion'
)
# Iteratively shrink the peaks until we have two peaks that we
# can use to separate the clump.
while True:
tmp = mh.morph.open(mh.morph.erode(peaks))
n = mh.label(tmp)[1]
if n == 2 or n == 0:
if n == 2:
peaks = tmp
break
peaks = tmp
# Select the two biggest peaks, since we want only two objects.
peaks = mh.label(peaks)[0]
sizes = mh.labeled.labeled_size(peaks)
index = np.argsort(sizes)[::-1][1:3]
for label in np.unique(peaks):
if label not in index:
peaks[peaks == label] = 0
peaks = mh.labeled.relabel(peaks)[0]
regions = mh.cwatershed(np.invert(dist), peaks)
# Use the line separating watershed regions to make the cut
se = np.ones((3,3), np.bool)
line = mh.labeled.borders(regions, Bc=se)
line[~obj_image] = 0
line = mh.morph.dilate(line)
# Ensure that cut is reasonable given user-defined criteria
test_cut_image = obj_image.copy()
test_cut_image[line] = False
subobjects, n_subobjects = mh.label(test_cut_image)
sizes = mh.labeled.labeled_size(subobjects)
smaller_object_area = np.min(sizes)
smaller_id = np.where(sizes == smaller_object_area)[0][0]
smaller_object = subobjects == smaller_id
do_cut = (
(smaller_object_area > min_cut_area) &
(np.sum(line) > 0)
)
if do_cut:
logger.debug('cut object #%d', oid)
y, x = np.where(line)
y_offset, x_offset = bboxes[oid][[0, 2]] - pad - 1
y += y_offset
x += x_offset
label_image[y, x] = 0
separated_image[y, x] = 0
else:
logger.debug('don\'t cut object #%d', oid)
mh.labeled.remove_regions(label_image, oid, inplace=True)
return mh.label(separated_image)[0]
def separate_clumped_objects(clumps_image, min_cut_area, min_area, max_area,
max_circularity, max_convexity):
'''Separates objects in `clumps_image` based on morphological criteria.
Parameters
----------
clumps_image: numpy.ndarray[Union[numpy.int32, numpy.bool]]
objects that should be separated
min_cut_area: int
minimal area an object must have (prevents cuts that would result
in too small objects)
min_area: int
minimal area an object must have to be considered a clump
max_area: int
maximal area an object can have to be considered a clump
max_circularity: float
maximal circularity an object must have to be considerd a clump
max_convexity: float
maximal convexity an object must have to be considerd a clump
Returns
-------
numpy.ndarray[numpy.uint32]
separated objects
See also
--------
:class:`jtlib.features.Morphology`
'''
logger.info('separate clumped objects')
label_image, n_objects = mh.label(clumps_image)
if n_objects == 0:
logger.debug('no objects')
return label_image
pad = 1
cutting_pass = 1
separated_image = label_image.copy()
while True:
logger.info('cutting pass #%d', cutting_pass)
cutting_pass += 1
label_image = mh.label(label_image > 0)[0]
f = Morphology(label_image)
values = f.extract()
index = (
(min_area < values['Morphology_Area']) &
(values['Morphology_Area'] <= max_area) &
(values['Morphology_Convexity'] <= max_convexity) &
(values['Morphology_Circularity'] <= max_circularity)
)
clumped_ids = values[index].index.values
not_clumped_ids = values[~index].index.values
if len(clumped_ids) == 0:
logger.debug('no more clumped objects')
break
mh.labeled.remove_regions(label_image, not_clumped_ids, inplace=True)
mh.labeled.relabel(label_image, inplace=True)
bboxes = mh.labeled.bbox(label_image)
for oid in np.unique(label_image[label_image > 0]):
bbox = bboxes[oid]
logger.debug('process clumped object #%d', oid)
obj_image = extract_bbox(label_image, bboxes[oid], pad=pad)
obj_image = obj_image == oid
# Rescale distance intensities to make them independent of clump size
dist = mh.stretch(mh.distance(obj_image))
# Find peaks that can be used as seeds for the watershed transform
thresh = mh.otsu(dist)
peaks = dist > thresh
n = mh.label(peaks)[1]
if n == 1:
logger.debug(
'only one peak detected - perform iterative erosion'
)
# Iteratively shrink the peaks until we have two peaks that we
# can use to separate the clump.
while True:
tmp = mh.morph.open(mh.morph.erode(peaks))
n = mh.label(tmp)[1]
if n == 2 or n == 0:
if n == 2:
peaks = tmp
break
peaks = tmp
# Select the two biggest peaks, since we want only two objects.
peaks = mh.label(peaks)[0]
sizes = mh.labeled.labeled_size(peaks)
index = np.argsort(sizes)[::-1][1:3]
for label in np.unique(peaks):
if label not in index:
peaks[peaks == label] = 0
peaks = mh.labeled.relabel(peaks)[0]
regions = mh.cwatershed(np.invert(dist), peaks)
# Use the line separating watershed regions to make the cut
se = np.ones((3,3), np.bool)
line = mh.labeled.borders(regions, Bc=se)
line[~obj_image] = 0
line = mh.morph.dilate(line)
# Ensure that cut is reasonable given user-defined criteria
test_cut_image = obj_image.copy()
test_cut_image[line] = False
subobjects, n_subobjects = mh.label(test_cut_image)
sizes = mh.labeled.labeled_size(subobjects)
smaller_object_area = np.min(sizes)
smaller_id = np.where(sizes == smaller_object_area)[0][0]
smaller_object = subobjects == smaller_id
do_cut = (
(smaller_object_area > min_cut_area) &
(np.sum(line) > 0)
)
if do_cut:
logger.debug('cut object #%d', oid)
y, x = np.where(line)
y_offset, x_offset = bboxes[oid][[0, 2]] - pad - 1
y += y_offset
x += x_offset
label_image[y, x] = 0
separated_image[y, x] = 0
else:
logger.debug('don\'t cut object #%d', oid)
mh.labeled.remove_regions(label_image, oid, inplace=True)
return mh.label(separated_image)[0]
def test_uint8():
# This did not work correctly in 0.9.5
a8 = np.zeros((5, 5), dtype=np.uint8)
ab = np.zeros((5, 5), dtype=bool)
assert np.all(distance(a8) == distance(ab))
def passmsg(self,child_part,parent_part):
'''function [score,Ix,Iy,Ik] = passmsg(child,parent)'''
INF = 1e+10
K , = child_part['filterid'].squeeze().shape
Nk,Ny,Nx = parent_part['score'].shape
Ix0 = np.zeros([K,Ny,Nx])
Iy0 = np.zeros([K,Ny,Nx])
score0 = np.zeros([K,Ny,Nx],dtype='float64') - INF
L = parent_part['filterid'].size
for k in range(K):
#assert child_part['w'][0,k]>0
#assert child_part['w'][2,k]>0
#print child_part['w'][0,k],child_part['w'][1,k],child_part['w'][2,k]
#score_tmp,Ix_tmp,Iy_tmp = self.dt2d(child_part['score'][k,:,:].astype('float64'),child_part['w'][0,k],child_part['w'][1,k],child_part['w'][2,k],child_part['w'][3,k])
#print child_part['score'][k,:,:].flags
score_tmp,Ix_tmp,Iy_tmp=mahotas.distance(np.asfortranarray(child_part['score'][k,:,:]),child_part['w'][0,k],child_part['w'][1,k],child_part['w'][2,k],child_part['w'][3,k])
startx = np.int_(child_part['startx'][k])-1
starty = np.int_(child_part['starty'][k])-1
step = child_part['step'].squeeze()
#% ending points
endy = starty+step*(Ny-1)+1
endx = startx+step*(Nx-1)+1
endy = min(child_part['score'].shape[1],endy)
endx = min(child_part['score'].shape[2],endx)
#y sample points
iy = np.arange(starty,endy,step).astype(int)
ix = np.arange(startx,endx,step).astype(int)
#print 'size iy,ix', iy.size,ix.size
if iy.size == 0 or ix.size ==0:
score = np.zeros([L,Ny,Nx],dtype='float64')
Ix = np.zeros([L,Ny,Nx]).astype(int)
Iy = np.zeros([L,Ny,Nx]).astype(int)
Ik = np.zeros([L,Ny,Nx]).astype(int)
return score,Ix,Iy,Ik
oy = sum(iy<0)
#print 'iy',iy
iy = iy[iy>=0]
ox = sum(ix<0)
ix = ix[ix>=0]
#sample scores
sp = score_tmp[iy,:][:,ix]
sx = Ix_tmp[iy,:][:,ix]
sy = Iy_tmp[iy,:][:,ix]
sz = sp.shape
#define msgs
iy = np.arange(oy,oy+sz[0])
ix = np.arange(ox,ox+sz[1])
iyp = np.tile(iy,[ix.shape[0],1]).T
ixp = np.tile(ix,[iy.shape[0],1])
score0[k,iyp,ixp] = sp
Ix0[k,iyp,ixp] = sx;
Iy0[k,iyp,ixp] = sy;
#% At each parent location, for each parent mixture 1:L, compute best child mixture 1:K
N = Nx*Ny;
i0 = np.arange(N).reshape(Ny,Nx)
i0row,i0col = np.unravel_index(i0, [Ny,Nx])
score = np.zeros([L,Ny,Nx],dtype='float64')
Ix = np.zeros([L,Ny,Nx]).astype(int)
Iy = np.zeros([L,Ny,Nx]).astype(int)
Ik = np.zeros([L,Ny,Nx]).astype(int)
for l in range(L):
b = child_part['b'][0,l,:]
score0b = (score0.transpose([1,2,0])+b).transpose([2,0,1])
score[l,:,:]= score0b.max(axis=0)
I = score0b.argmax(axis=0)
#print Ix.shape, Ix0.shape, i0row.shape,i0col.shape,I.shape
Ix[l,:,:] = Ix0[I,i0row,i0col]
Iy[l,:,:] = Iy0[I,i0row,i0col]
Ik[l,:,:] = I
return score,Ix,Iy,Ik
from scipy import ndimage
import scipy
# import numpy for standard numerical calculations
import numpy as np
# read the image with mahotas as a grey image
img=m.imread('./testimg4.jpg',as_grey=True)
# read the image with mahotas again to obtain a color image where we can draw the ReebGraph in red (vertices) and green (edges)
imgColor=m.imread('./testimg4.jpg')
# Threshhold to remove artifacts from the jpg compression
img=(img>100)
#get the dimensions of the image
x,y = np.shape(img)
#use the distance transform to obtain the distances per pixel of the medial axis
dmap = m.distance(img,metric='manhatten')
#use mathamatical morphology to obtain the medial axis (thinning function of mahotas)
skelImg=m.thin(img)
# draw the medial axis in the image
for idx,i in enumerate(skelImg):
for jdx,j in enumerate(i):
if skelImg[idx,jdx]==True:
imgColor[idx,jdx]=(255,1,1)
try:
imgColor[idx+1,jdx]=(255,1,1)
except:
pass
imgColor[idx-1,jdx]=(255,1,1)
try:
imgColor[idx,jdx+1]=(255,1,1)
filtered = mh.gaussian_filter(image, 10) # Koristimo gauss filter za izostravanje slike.
result = filtered.astype('uint8') # Funckija mh.gaussian_filter() postavlja vrijednost varijable u float64.
rmax = mh.regmax(result) # Trazi maksimalne vrijednosti.
print rmax
pylab.imshow(mh.overlay(image, rmax))
pylab.show()
labeled, nr_objects = mh.label(rmax)
# mh.overlay() funckija postavlja img kao pozadinu a preko nje postavlja vrijednosti variable rmax u crvenom kanalu.
print ('Broj pronadenih objekata je {}.'.format(nr_objects))
pylab.imshow(labeled)
# pylab.gray()
pylab.show()
dist = mh.distance(result)
dist = dist.max() - dist
dist -= dist.min()
dist = dist/float(dist.ptp()) * 255
dist = dist.astype(np.uint8)
objects = mh.cwatershed(dist, labeled)
whole = mh.segmentation.gvoronoi(objects) # Voronoi segemtacija, svaki piksel poprima vrijednost najblizeg maksimuma.
pylab.imshow(objects)
pylab.show()
# print (labeled).shape
# print (labeled).max()
# print (labeled).min()
from __future__ import print_function import pylab as p import numpy as np import mahotas f = np.ones((256, 256), bool) f[200:, 240:] = False f[128:144, 32:48] = False # f is basically True with the exception of two islands: one in the lower-right # corner, another, middle-left dmap = mahotas.distance(f) p.imshow(dmap) p.show()
sizes = mahotas.labeled.labeled_size(colonies)
# print(sizes)
too_small = np.where(sizes < 100)
colonies = mahotas.labeled.remove_regions(colonies, too_small)
#colonies = mahotas.labeled.remove_bordering(colonies)
colonies, n_colonies = mahotas.labeled.relabel(colonies)
print('Found {} colonies.'.format(n_colonies))
# plt.imshow(colonies)
# print(colonies)
# Investigate nuclei within cell clusters
# Now, we compute the distance transform:
distances = mahotas.stretch(mahotas.distance(local_threshed))
# We find and label the regional maxima:
Bc = np.ones((9,9))
maxima = mahotas.morph.regmax(distances, Bc=Bc)
spots,n_spots = mahotas.label(maxima, Bc=Bc)
print('Found {} maxima.'.format(n_spots))
# plt.imshow(spots)
# Finally, to obtain the image above, we invert the distance transform
# (because of the way that cwatershed is defined) and compute the watershed:
surface = (distances.max() - distances)
areas = mahotas.cwatershed(surface, spots)
areas *= local_threshed
import numpy as np
import mahotas
from pylab import imshow, savefig
A = np.zeros((100,100), bool)
A[40:60] = 1
W = mahotas.thin(A)
D = mahotas.distance(~W)
imshow(D)
savefig('distance.png')
#load segmentation results
labels = []
for i, row in df.reset_index().iterrows():
label = io.imread(analysis_dir + 'segm/' + row['Filename'])
labels.append(label)
labels = np.array(labels)
segs = labels > 0
#Foward and backward
k = 0
while (k < 2 * (repeat + 1)):
for i in range(df.shape[0] - N):
intersect = np.prod(segs[i:i + N], axis=0)
intersect_l = measure.label(intersect)
distances = mh.distance(segs[i])
surface = (distances.max() - distances)
areas = mh.cwatershed(surface, intersect_l)
new_label = areas * segs[i]
labels[i] = new_label
segs[i] = new_label > 0
labels = labels[::-1]
segs = segs[::-1]
k += 1
#Save refined segmentations
for i, row in df.reset_index().iterrows():
io.imsave(analysis_dir + 'segm_refined/' + row['Filename'],
def detectMyotube(self,
segmentationMethod='seg1',
sizeThresh=0,
tophat=True,
tophatImgList=[]):
if tophat == True and tophatImgList == []:
tophatImgList = self.tophatAllPlanes()
elif tophat == True and tophatImgList != []:
#tophatImgList = tophatImgList[tophatImgList.keys()[0]]
tophatImgList = tophatImgList[0]
elif tophat == False:
tophatImgList = self.images
# median -> histeq -> otsu -> segmentation (histeq and otsu by region)
if segmentationMethod == 'seg1':
img_mip = self.maximumIntensityProjection(tophatImgList)
img_median = self.smooth(img_mip)
img_histeq, _ = imtools.histeq(img_median)
T = otsu(np.uint(img_histeq), ignore_zeros=True)
img_bin = (np.uint(img_histeq) > T)
img_labeled, _ = label(img_bin)
markers, counts = self.segmentTubes1(img_labeled, img_histeq)
# segmentation by watershed
img_labeled = img_bin * cwatershed(-distance(img_bin), markers)
result = {
'MIP': img_mip,
'median': img_median,
'histEq': img_histeq,
'otsu': T,
'bin': img_bin
}
# median -> otsu -> segmentation (histeq and otsu by region)
elif segmentationMethod == 'seg2':
img_mip = self.maximumIntensityProjection(tophatImgList)
img_median = self.smooth(img_mip)
T = otsu(np.uint(img_median), ignore_zeros=True)
img_bin = (np.uint(img_median) > T)
img_labeled, _ = label(img_bin)
markers, counts = self.segmentTubes2(img_labeled, img_median)
# segmentation by watershed
img_labeled = img_bin * cwatershed(-distance(img_bin), markers)
result = {
'MIP': img_mip,
'median': img_median,
'otsu': T,
'bin': img_bin
}
# median -> histeq -> otsu -> segmentation (histeq and cut regions less than mean-intensity by region)
elif segmentationMethod == 'seg3':
img_mip = self.maximumIntensityProjection(tophatImgList)
img_median = self.smooth(img_mip)
img_histeq, _ = imtools.histeq(img_median)
T = otsu(np.uint(img_histeq), ignore_zeros=True)
img_bin = (np.uint(img_histeq) > T)
img_labeled, _ = label(img_bin)
markers, counts = self.segmentTubes3(img_labeled, img_histeq)
# segmentation by watershed
img_labeled = img_bin * cwatershed(-distance(img_bin), markers)
result = {
'MIP': img_mip,
'median': img_median,
'histEq': img_histeq,
'otsu': T,
'bin': img_bin
}
# median -> histeq -> otsu -> segmentation (cut regions less than mean-intensity by region)
elif segmentationMethod == 'seg4':
img_mip = self.maximumIntensityProjection(tophatImgList)
img_median = self.smooth(img_mip)
img_histeq, _ = imtools.histeq(img_median)
T = otsu(np.uint(img_histeq), ignore_zeros=True)
img_bin = (np.uint(img_histeq) > T)
img_labeled, _ = label(img_bin)
markers, counts = self.segmentTubes4(img_labeled, img_histeq)
# segmentation by watershed
img_labeled = img_bin * cwatershed(-distance(img_bin), markers)
result = {
'MIP': img_mip,
'median': img_median,
'histEq': img_histeq,
'otsu': T,
'bin': img_bin
}
# median -> otsu -> segmentation (cut regions less than mean-intensity by region)
elif segmentationMethod == 'seg5':
img_mip = self.maximumIntensityProjection(tophatImgList)
img_median = self.smooth(img_mip)
T = otsu(np.uint(img_median), ignore_zeros=True)
img_bin = (np.uint(img_median) > T)
img_labeled, _ = label(img_bin)
markers, counts = self.segmentTubes4(img_labeled, img_median)
# segmentation by watershed
img_labeled = img_bin * cwatershed(-distance(img_bin), markers)
result = {
'MIP': img_mip,
'median': img_median,
'otsu': T,
'bin': img_bin
}
# non-segmentation
else:
img_mip = self.maximumIntensityProjection(tophatImgList)
img_median = self.smooth(img_mip)
img_histeq, _ = imtools.histeq(img_median)
T = otsu(np.uint(img_histeq))
img_bin = (np.uint(img_histeq) > T)
img_labeled, counts = label(img_bin)
result = {
'MIP': img_mip,
'median': img_median,
'histEq': img_histeq,
'otsu': T,
'bin': img_bin
}
print('non-segmentation')
print('Otsu\'s threshold:', T)
print('Found {} objects.'.format(counts))
sizes = labeled_size(img_labeled)
img_labeled = remove_regions_where(
img_labeled, sizes < sizeThresh) #origin 10000, triangle 8585
img_relabeled, counts = relabel(img_labeled)
result['label'] = img_relabeled
result['count'] = counts
print('After filtering and relabeling, there are {} objects left.'.
format(counts))
result['labeledSkeleton'] = self.labeledSkeleton(img_relabeled)
return ProcessImages(result)
def compare_slow(bw):
for metric in ('euclidean', 'euclidean2'):
f = distance(bw, metric)
sd = _slow_dist(bw, metric)
assert np.all(f == sd)
def extract(self):
'''Extracts point pattern features.
Returns
-------
pandas.DataFrame
extracted feature values for each object in
:attr:`label_image <jtlib.features.PointPattern.label_image>`
'''
logger.info('extract point pattern features')
features = dict()
for obj in self.parent_object_ids:
parent_obj_img = self.get_parent_object_mask_image(obj)
points_img = self.get_points_object_label_image(obj)
point_ids = np.unique(points_img)[1:]
mh.labeled.relabel(points_img, inplace=True)
size = np.sum(parent_obj_img)
abs_border_dist_img = mh.distance(parent_obj_img).astype(float)
rel_border_dist_img = abs_border_dist_img / size
centroids = mh.center_of_mass(points_img, labels=points_img)
centroids = centroids[1:, :].astype(int)
indexer = np.arange(centroids.shape[0])
if len(indexer) == 0:
continue
if len(indexer) == 1:
y, x = centroids[0, :]
values = [
abs_border_dist_img[y, x],
rel_border_dist_img[y, x],
np.nan,
np.nan,
np.nan,
np.nan,
np.nan,
np.nan
]
features[point_ids[0]] = values
continue
for i, c in enumerate(centroids):
abs_distances = cdist([c], centroids)[0, :]
rel_distances = abs_distances / size
idx = indexer != i
y, x = c
values = [
abs_border_dist_img[y, x],
rel_border_dist_img[y, x],
np.nanmin(abs_distances[idx]),
np.nanmin(rel_distances[idx]),
np.nanmean(abs_distances[idx]),
np.nanmean(rel_distances[idx]),
np.nanstd(abs_distances[idx]),
np.nanstd(rel_distances[idx]),
]
features[point_ids[i]] = values
ids = features.keys()
values = list()
nans = [np.nan for _ in range(len(self.names))]
for i in self.object_ids:
if i not in ids:
logger.warn('values missing for object #%d', i)
features[i] = nans
values.append(features[i])
return pd.DataFrame(values, columns=self.names, index=self.object_ids)
def region_growing(labelImg):
distances = mahotas.stretch(mahotas.distance(labelImg > 0))
surface = numpy.int32(distances.max() - distances)
areas = mahotas.cwatershed(surface, labelImg)
return areas
print(nr_objects)
plt.imshow(labeled)
plt.jet()
@interact(sigma=(1.,16.))
def check_sigma(sigma):
dnaf = mh.gaussian_filter(dna.astype(float), sigma)
maxima = mh.regmax(mh.stretch(dnaf))
maxima = mh.dilate(maxima, np.ones((5,5)))
plt.imshow(mh.as_rgb(np.maximum(255*maxima, dnaf), dnaf, dna > T_mean))
sigma = 12.0
dnaf = mh.gaussian_filter(dna.astype(float),sigma)
maxima = mh.regmax(mh.stretch(dnaf))
maxima,_= mh.label(maxima)
plt.imshow(maxima)
dist = mh.distance(bin_image)
plt.imshow(dist)
dist = 255 - mh.stretch(dist)
watershed = mh.cwatershed(dist,maxima)
plt.imshow(watershed)
watershed *= bin_image
plt.imshow(watershed)
watershed = mh.labeled.remove_bordering(watershed)
plt.imshow(watershed)
sizes = mh.labeled.labeled_size(watershed)
# The conversion below is not necessary in newer versions of mahotas: watershed = watershed.astype(np.intc)
@interact(min_size=(100,4000,20))
def do_plot(min_size):
filtered = mh.labeled.remove_regions_where(watershed, sizes < min_size)
print("filtering {}...".format(min_size))
plt.imshow(filtered)
import mahotas
import numpy as np
from matplotlib import pyplot as plt
import random
from matplotlib import colors as c
nuclear = mahotas.imread('data/flower.png')
nuclear = nuclear[:, :, 0]
nuclear = mahotas.gaussian_filter(nuclear, 1.)
threshed = (nuclear > nuclear.mean())
distances = mahotas.stretch(mahotas.distance(threshed))
Bc = np.ones((9, 9))
maxima = mahotas.morph.regmax(distances, Bc=Bc)
spots, n_spots = mahotas.label(maxima, Bc=Bc)
surface = (distances.max() - distances)
areas = mahotas.cwatershed(surface, spots)
areas *= threshed
plt.jet()
rmap = c.ListedColormap(np.random.rand(256, 3))
plt.imshow(areas, cmap=rmap)
plt.show()
colonies, n_colonies = mahotas.label(threshed)
sizes = mahotas.labeled.labeled_size(colonies)
# print(sizes)
too_small = np.where(sizes < 100)
colonies = mahotas.labeled.remove_regions(colonies, too_small)
#colonies = mahotas.labeled.remove_bordering(colonies)
colonies, n_colonies = mahotas.labeled.relabel(colonies)
print('Found {} colonies.'.format(n_colonies))
# plt.imshow(colonies)
# print(colonies)
# Investigate nuclei within cell clusters
# Now, we compute the distance transform:
distances = mahotas.stretch(mahotas.distance(local_threshed))
# We find and label the regional maxima:
Bc = np.ones((9, 9))
maxima = mahotas.morph.regmax(distances, Bc=Bc)
spots, n_spots = mahotas.label(maxima, Bc=Bc)
print('Found {} maxima.'.format(n_spots))
# plt.imshow(spots)
# Finally, to obtain the image above, we invert the distance transform
# (because of the way that cwatershed is defined) and compute the watershed:
surface = (distances.max() - distances)
areas = mahotas.cwatershed(surface, spots)
areas *= local_threshed
def keypress(event):
global nodes, edges, node_labels, G, root, select
print('press', event.key)
sys.stdout.flush()
if event.key == 'x': # delete closest branch
e = findClosestEdge(G, (event.xdata, event.ydata))
undo_stack.append((G.copy(), root, Cercle))
G.remove_edge(*e)
updateMeasures(G, root)
plot_graph(G)
fig.canvas.draw()
if event.key == 'u': # undo
if len(undo_stack) > 0:
print('Undo')
G, root, C = undo_stack.pop()
setRoot(root)
updateMeasures(G, root)
updateCircles(G, C)
plot_graph(G)
fig.canvas.draw()
else:
print('No further undo')
if event.key == 'r': #re-build the graph from skeleton
print('Rebuilding tree')
undo_stack.append((G.copy(), root, Cercle))
G = buildGraph(img, skel)
plot_graph(G)
updateCircles(G, Cercle)
fig.canvas.draw()
if event.key == 'n': #add node
p = findClosestSkel(skel2, (event.xdata, event.ydata))
print('closest node is (%5.1f, %5.1f)' % p)
undo_stack.append((G.copy(), root, Cercle))
print('adding node')
try:
addNodeSkel(G, p, skel2, j, t, dmap)
except:
print(' distance transform...')
dmap = mh.distance(img)
addNodeSkel(G, p, skel2, j, t, dmap)
Cercle[p] = plt.Circle(p,
radius=(G.node[p]['dia'] / 2),
alpha=terminal_disk_alpha,
color=terminal_disk_color)
updateCircles(G, Cercle)
updateMeasures(G, root)
report(G)
plot_graph(G)
fig.canvas.draw()
if select == None: # verify the selection
print('No active selection, please select a node')
return
if event.key == 't': # change the root
undo_stack.append((G.copy(), root, Cercle))
root = select
print('New root: ' + str(root))
setRoot(root)
updateMeasures(G, root)
report(G)
plot_graph(G)
fig.canvas.draw()
if event.key == 'd': # delete closest node
p = select
if p == root:
print('Cannot remove the root.')
return
undo_stack.append((G.copy(), root, Cercle))
deleteNode(G, p)
Cercle[p].remove()
#report(G)
updateMeasures(G, root)
plot_graph(G)
fig.canvas.draw()
if event.key == 'a': #marking of fertile nodes
p = select
print('this node is now fertile')
undo_stack.append((G.copy(), root, Cercle))
G.node[p]['node_color'] = node_color_fertile
G.node[p]['fertile'] = True
G[p][G.node[p]['parent']]['fertile'] = True
report(G)
plot_graph(G)
fig.canvas.draw()
if event.key == 'alt+e': #hide any latest selection
print('hide caracteristics')
undo_stack.append((G.copy(), root, Cercle))
for p in G:
if G.node[p]['node_color'] != node_color:
G.node[p]['node_color'] = node_color
plot_graph(G)
fig.canvas.draw()
report(G)
if event.key == 'b': # diameter modification
p = select
print('closest node is (%5.1f, %5.1f)' % p)
undo_stack.append((G.copy(), root, Cercle))
G.node[p]['dia_2'] = 2 * dist(p, (event.xdata, event.ydata))
updateMeasures(G, root)
Cercle[p].remove()
Cercle[p] = plt.Circle(p,
radius=G.node[p]['dia_2'] / 2,
alpha=terminal_disk_alpha,
color=terminal_disk_color)
plt.gca().add_patch(Cercle[p])
plot_graph(G)
fig.canvas.draw()
def generate_image(cells, shape, max_dist=5):
thetas = 360 * np.random.rand(len(cells))
data_list = [cell.data.rotate(theta) for cell, theta in zip(cells, thetas)]
assert all([data.names == data_list[0].names for data in data_list
]), 'All cells must have the same data elements'
out_dict = {
name: np.zeros(shape)
for name, dclass in zip(data_list[0].names, data_list[0].dclasses)
if dclass != 'storm'
}
for i, data in enumerate(data_list):
valid_position = False
while not valid_position:
pos_x = int(np.round(shape[1] * np.random.rand()))
pos_y = int(np.round(shape[0] * np.random.rand()))
min1 = pos_y - int(np.floor(data.shape[0]))
max1 = min1 + data.shape[0]
min2 = pos_x - int(np.floor(data.shape[1]))
max2 = min2 + data.shape[1]
# Crop the data for when the cell is on the border of the image
d_min1 = np.max([0 - min1, 0])
d_max1 = np.min(
[data.shape[0] + (shape[0] - pos_y), data.shape[0]])
d_min2 = np.max([0 - min2, 0])
d_max2 = np.min(
[data.shape[1] + (shape[1] - pos_x), data.shape[1]])
data_cropped = data[d_min1:d_max1, d_min2:d_max2]
# Limit image position to the edges of the image
min1 = np.max([min1, 0])
max1 = np.min([max1, shape[0]])
min2 = np.max([min2, 0])
max2 = np.min([max2, shape[1]])
temp_binary = np.zeros(shape)
temp_binary[min1:max1, min2:max2] = data_cropped.binary_img
out_binary = (out_dict['binary'] > 0).astype(int)
distance_map = mh.distance(1 - out_binary, metric='euclidean')
if np.any(distance_map[temp_binary.astype(bool)] < max_dist):
continue
valid_position = True
for name in data.names:
data_elem = data_cropped.data_dict[name]
if data_elem.dclass == 'storm':
data_elem['x'] += min2
data_elem['y'] += min1
xmax, ymax = shape[1], shape[0]
bools = (data_elem['x'] < 0) + (data_elem['x'] > xmax) + (
data_elem['y'] < 0) + (data_elem['y'] > ymax)
data_out = data_elem[~bools].copy()
if name in out_dict:
out_dict[name] = np.append(out_dict[name], data_out)
else:
out_dict[name] = data_out
continue
elif data_elem.dclass == 'binary':
out_dict[name][min1:max1, min2:max2] += ((i + 1) * data_elem)
else:
out_dict[name][min1:max1, min2:max2] += data_elem
return out_dict
import numpy as np
from matplotlib import pyplot as plt
try:
nuclear_path = path.join(
path.dirname(path.abspath(__file__)),
'data',
'nuclear.png')
except NameError:
nuclear_path = path.join('data', 'nuclear.png')
nuclear = mahotas.imread(nuclear_path)
nuclear = nuclear[:,:,0]
nuclear = mahotas.gaussian_filter(nuclear, 1.)
threshed = (nuclear > nuclear.mean())
distances = mahotas.stretch(mahotas.distance(threshed))
Bc = np.ones((9,9))
maxima = mahotas.morph.regmax(distances, Bc=Bc)
spots,n_spots = mahotas.label(maxima, Bc=Bc)
surface = (distances.max() - distances)
areas = mahotas.cwatershed(surface, spots)
areas *= threshed
import random
from matplotlib import colors as c
colors = map(plt.cm.jet,range(0, 256, 4))
random.shuffle(colors)
colors[0] = (0.,0.,0.,1.)
def region_growing(labelImg):
distances = mahotas.stretch(mahotas.distance(labelImg>0))
surface = numpy.int32(distances.max() - distances)
areas = mahotas.cwatershed(surface, labelImg)
return areas
def test_uint8():
# This did not work correctly in 0.9.5
a8 = np.zeros((5,5), dtype=np.uint8)
ab = np.zeros((5,5), dtype=bool)
assert np.all(distance(a8) == distance(ab))
if __name__ == '__main__':
pylab.axis('off')
img = cv2.imread('img2.png')
img = cv2.cvtColor(img, cv2.cv.CV_BGR2GRAY)
img /= 255
# Rutovitz
rut = nc_rutovitz(img)
pylab.imshow(rut, interpolation='nearest')
pylab.savefig('rutovitz.png', bbox_inches='tight')
# Yokoi 4
yokoi4 = nc_yokoi(img, 4)
pylab.imshow(yokoi4, interpolation='nearest')
pylab.savefig('yokoi4.png', bbox_inches='tight')
# Yokoi 8
yokoi8 = nc_yokoi(img, 8)
pylab.imshow(yokoi8, interpolation='nearest')
pylab.savefig('yokoi8.png', bbox_inches='tight')
# Transformada de distancia
img = cv2.imread('img.png')
img = cv2.cvtColor(img, cv2.cv.CV_BGR2GRAY)
dist = mahotas.distance(img, metric='euclidean')
pylab.imshow(dist, interpolation='nearest')
pylab.gray()
pylab.savefig('dist.png', bbox_inches='tight')
