def remove_abnormal_samples(seg, sigma=0.9):
'''removes abnormal samples based on sig. deviation from mean centroid,
then removes based on mean size'''
# Get centroids of samples
centroids = []
for idx, lab in enumerate(measure.regionprops(scipy.ndimage.label(seg)[0])):
centroids.append(lab.centroid)
row_vals = [x[0] for x in centroids]
# Get relevant stats of row values
mean = np.mean(row_vals)
std = np.std(row_vals)
hi = mean + sigma*std
lo = mean - sigma*std
# Eliminate sig. deviation from mean
for idx, lab in enumerate(measure.regionprops(scipy.ndimage.label(seg)[0])):
centroid = lab.centroid
if centroid[0] < lo or centroid[0] > hi:
seg = floodfill_fast(
seg,
x=int(centroid[0]),
y=int(centroid[1]),
value=0,
border_color=0,
dtype=np.uint8
)[0]
# Get sizes of samples
areas = []
for idx, lab in enumerate(measure.regionprops(scipy.ndimage.label(seg)[0])):
areas.append(lab.filled_area)
mean = np.mean(areas)
std = np.std(areas)
hi = mean + 3*sigma*std
lo = mean - 3*sigma*std
# Eliminate sig. deviation from mean
for idx, lab in enumerate(measure.regionprops(scipy.ndimage.label(seg)[0])):
area = lab.filled_area
centroid = lab.centroid
if area < lo or area > hi:
seg = floodfill_fast(
seg,
x=int(centroid[0]),
y=int(centroid[1]),
value=0,
border_color=0,
dtype=np.uint8
)[0]
return seg
def find_standards(seg, n_standards=6):
'''finds standards based on regular proximity across row
only actually requires seg
returns new image with standards == 2, samples == 1
'''
best_std_candidate_row = -1
previous_dists = []
previous_mean = -1
previous_std = np.inf
####REWRITE TO REDUCE NSTDS-! BY ONE EACH TIME IF DOESNT FIND BESTSTDCANDIDATEROW
cutoff = n_standards - 1 # ideally, want to find this many gaps
while True:
if cutoff == 0:
raise ValueError('Error: Did not find any standards')
break
for i, row in enumerate(seg):
dists = object_distances(row)
if len(dists) >= cutoff and dists != previous_dists:
mean = np.mean(dists)
std = np.std(dists)
if std < previous_std:
previous_mean = mean
previous_std = std
best_std_candidate_row = i
previous_dists = dists
if best_std_candidate_row == -1:
print('Failed to detect all 6 standards, will now try to find some')
cutoff -= 1
else:
break
if best_std_candidate_row == -1:
raise ValueError('unable to identify all 6 standards, take new image')
#plot_row(seg, best_std_candidate_row)
# Now, we have row with stds. Lets color the standards:
best = best_std_candidate_row
new = seg.copy()
# get list of ys marking each standard
lab = scipy.ndimage.label(seg[best])[0]
ys = []
for i in range(1, lab.max()+1):
for idx, item in enumerate(lab):
if item == i:
ys.append(idx)
break
print(ys)
for y in ys:
new = floodfill_fast(new, x=best, y=y, value=2, border_color=0, dtype=np.uint8)[0]
return new
