def morph_snakes(data,
mask,
slice=None,
scale=0.5,
alpha=1000,
sigma=1,
smoothing_ls=1,
threshold=0.3,
balloon=1,
max_iters=50,
show=False,
show_now=True):
data_o = data.copy()
mask_o = mask.copy()
if scale != 1:
# data = skitra.rescale(data, scale=scale, preserve_range=True).astype(np.uint8)
# mask = skitra.rescale(mask, scale=scale, preserve_range=True).astype(np.bool)
data = tools.resize3D(data, scale, sliceId=0)
mask = tools.resize3D(mask, scale, sliceId=0)
gI = morphsnakes.gborders(data, alpha=alpha, sigma=sigma)
# Morphological GAC. Initialization of the level-set.
mgac = morphsnakes.MorphGAC(gI,
smoothing=smoothing_ls,
threshold=threshold,
balloon=balloon)
mgac.levelset = mask
mgac.run(iterations=max_iters)
seg = mgac.levelset
if scale != 1:
# data = tools.resize_ND(data, shape=orig_shape)
# mask = tools.resize_ND(mask, shape=orig_shape)
seg = tools.resize_ND(seg, shape=data_o.shape)
if show:
tools.visualize_seg(data_o,
seg,
mask_o,
slice=slice,
title='morph snakes',
show_now=show_now)
return data, mask, seg
def lankton_ls(im,
mask,
method='sfm',
slice=None,
max_iters=1000,
rad=10,
alpha=0.1,
scale=1.,
show=False,
show_now=True):
if scale != 1:
# im = skitra.rescale(im, scale=scale, preserve_range=True).astype(np.uint8)
# mask = skitra.rescale(mask, scale=scale, preserve_range=True).astype(np.bool)
im = tools.resize_ND(im, scale=scale).astype(np.uint8)
mask = tools.resize_ND(mask, scale=scale).astype(np.bool)
seg = lankton_lls.run(
im,
mask,
method='sfm',
max_iter=1000,
rad=rad,
alpha=alpha,
# slice=slice, show=show, show_now=show_now)
slice=slice,
show=True,
show_now=True)
if show:
tools.visualize_seg(data,
mask,
seg,
slice=slice,
title='morph snakes',
show_now=show_now)
return im, mask, seg
min_diff = 1
alpha = 0.9
scale = 0.5
max_iters = 100
smoothing = True
show = True
show_now = False
_, seeds = tools.initialize_graycom(data, slice=slice_ind, show=show, show_now=show_now)
# seeds = tools.eroding3D(seeds, np.ones((3, 3))).astype(np.uint8)
seeds = tools.morph_ND(seeds, 'erosion', selem=np.ones((3, 3)), slicewise=True, sliceId=0).astype(np.uint8)
plt.figure()
plt.subplot(121), plt.imshow(data[slice_ind,...], 'gray')
plt.subplot(122), plt.imshow(seeds[slice_ind,...], 'jet')
plt.show()
lrg = LoReGro(data, seeds, ball_r_out=ball_r_out, ball_r=ball_r, min_diff=min_diff, alpha=alpha,
scale=scale, max_iters=max_iters, smoothing=smoothing, show=show, show_now=show_now)
lrg.run_segmentation()
seg = lrg.segmentation
# plt.figure()
# plt.suptitle('LoReGro segmentation')
# plt.subplot(131), plt.imshow(data, 'gray'), plt.title('input')
# plt.subplot(132), plt.imshow(seeds, 'jet'), plt.title('seeds')
# plt.subplot(133), plt.imshow(seg[0,...].astype(np.uint8), 'gray'), plt.title('segmentation')
# plt.show()
tools.visualize_seg(data, seg[0,...], mask=seeds, title='loregro')
def run_param_tuning_morphsnakes(im_in,
gt_mask,
slice_ind,
smoothing=True,
scale=0.5,
init_scale=0.5):
if smoothing:
im_in = tools.smoothing(im_in, sigmaSpace=10, sigmaColor=10, sliceId=0)
im = im_in.copy()
mask_init = tools.initialize_graycom(im,
slice,
distances=[
1,
],
scale=init_scale)
# parameters tuning ---------------------------------------------------------------
alpha_v = (10, 100, 500)
sigma_v = (1, 5)
threshold_v = (0.1, 0.5, 0.9)
balloon_v = (1, 2, 4)
method = 'morphsnakes'
# create workbook for saving the resulting pracision, recall and f-measure
workbook = xlsxwriter.Workbook(
'/home/tomas/Dropbox/Data/liver_segmentation/%s/%s.xlsx' %
(method, method))
worksheet = workbook.add_worksheet()
# Add a bold format to use to highlight cells.
bold = workbook.add_format({'bold': True})
# green = workbook.add_format({'bg_color': '#C6EFCE'})
worksheet.write(0, 0, 'F-MEASURE', bold)
worksheet.write(0, 1, 'PRECISION', bold)
worksheet.write(0, 2, 'RECALL', bold)
worksheet.write(0, 4, 'alpha', bold)
worksheet.write(0, 5, 'sigma', bold)
worksheet.write(0, 6, 'threshold', bold)
worksheet.write(0, 7, 'ballon', bold)
worksheet.write(0, 8, 'scale', bold)
worksheet.write(0, 9, 'scale_init', bold)
params_tune = []
for p in itertools.product(alpha_v, sigma_v, threshold_v, balloon_v):
params_tune.append(p)
# for i, (alpha, sigma, threshold, balloon) in enumerate(itertools.product(alpha_v, sigma_v, threshold_v, balloon_v)):
for i in range(len(params_tune)):
alpha, sigma, threshold, balloon = params_tune[i]
# threshold = 0.9
# balloon = 4
print '\n -- it #%i/%i --' % (i + 1, len(params_tune))
print 'Working with alpha=%i, sigma=%i, threshold=%.1f, balloon=%i' % (
alpha, sigma, threshold, balloon)
params = {
'alpha': alpha,
'sigma': sigma,
'smoothing_ls': 1,
'threshold': threshold,
'balloon': balloon,
'max_iters': 1000,
'scale': scale,
'init_scale': init_scale,
'smoothing': False,
'save_fig': False,
'show': True,
'show_now': False
}
im, mask, seg = run(im_in,
slice=slice_ind,
mask=mask_init,
method=method,
**params)
# print 'shapes 1: im={}, mask={}, seg={}'.format(im.shape, mask.shape, seg.shape)
if im.shape != im_in.shape:
im = tools.resize_ND(im, shape=im_in.shape)
if mask.shape != gt_mask.shape:
mask = tools.resize_ND(mask, shape=gt_mask.shape)
if seg.shape != gt_mask.shape:
seg = tools.resize_ND(seg, shape=gt_mask.shape)
# print 'shapes 2: im={}, mask={}, seg={}'.format(im.shape, mask.shape, seg.shape)
# calculating precision, recall and f_measure
precision, recall, f_measure = calc_statistics(seg, gt_mask)
# print 'shapes 3: im={}, mask={}, seg={}'.format(im.shape, mask.shape, seg.shape)
# saving data
datap = {'im': im, 'mask': mask, 'seg': seg, 'params': params}
dirname = '/home/tomas/Dropbox/Data/liver_segmentation/%s/' % method
dataname = '%s_al%i_si%i_th%s_ba%i.pklz' % (
method, alpha, sigma, str(threshold).replace('.', ''), balloon)
f = gzip.open(os.path.join(dirname, dataname), 'w')
pickle.dump(datap, f)
f.close()
# saving results
cols = [0, 1, 2, 4, 5, 6, 7, 8, 9]
items = (f_measure, precision, recall, alpha, sigma, threshold,
balloon, params['scale'], params['init_scale'])
for j, it in enumerate(items):
worksheet.write(i + 1, cols[j], it)
# creating visualization
fig = tools.visualize_seg(im,
seg,
mask,
slice=slice_ind,
title=method,
for_save=True,
show_now=False)
fname = '/home/tomas/Dropbox/Data/liver_segmentation/%s/%s_al%i_si%i_th%s_ba%i.png' \
% (method, method, alpha, sigma, str(threshold).replace('.', ''), balloon)
fig.savefig(fname)
plt.close('all')
workbook.close()
plt.subplot(121), plt.imshow(data[slice_ind, ...], 'gray')
plt.subplot(122), plt.imshow(seeds[slice_ind, ...], 'jet')
# plt.show()
lrg = LoReGro(data,
seeds,
ball_r_out=ball_r_out,
ball_r=ball_r,
min_diff=min_diff,
alpha=alpha,
scale=scale,
max_iters=max_iters,
smoothing=smoothing,
show=show,
show_now=show_now)
lrg.run_segmentation()
seg = lrg.segmentation
# plt.figure()
# plt.suptitle('LoReGro segmentation')
# plt.subplot(131), plt.imshow(data, 'gray'), plt.title('input')
# plt.subplot(132), plt.imshow(seeds, 'jet'), plt.title('seeds')
# plt.subplot(133), plt.imshow(seg[0,...].astype(np.uint8), 'gray'), plt.title('segmentation')
# plt.show()
# tools.visualize_seg(data, seg[0,...], mask=seeds, title='loregro')
tools.visualize_seg(data,
seg,
mask=seeds,
slice=slice_ind,
title='loregro')
def segmentation_stats(data_struc, scale=1, smoothing=True):
# fnames = load_data(datadir)
# fnames = ['/home/tomas/Dropbox/Data/medical/dataset/180_arterial-.pklz',]
# creating workbook
workbook = xlsxwriter.Workbook('/home/tomas/Dropbox/Data/liver_segmentation/final_alg/liver_seg_stats.xlsx')
worksheet = workbook.add_worksheet()
# Add a bold format to use to highlight cells.
bold = workbook.add_format({'bold': True})
# green = workbook.add_format({'bg_color': '#C6EFCE'})
col_gc_fmea = 3
col_gc_prec = 6
col_gc_rec = 9
col_ls_fmea = 4
col_ls_prec = 7
col_ls_rec = 10
cols_gc = (3, 6, 9)
cols_ls = (4, 7, 10)
start_row = 2
worksheet.write(0, 0, 'DATA', bold)
worksheet.write(0, 1, 'slice', bold)
worksheet.write(0, 3, 'F-MEASURE', bold)
worksheet.write(0, 6, 'PRECISION', bold)
worksheet.write(0, 9, 'RECALL', bold)
worksheet.write(1, col_gc_fmea, 'Grow Cut', bold)
worksheet.write(1, col_ls_fmea, 'Level Sets', bold)
worksheet.write(1, col_gc_prec, 'Grow Cut', bold)
worksheet.write(1, col_ls_prec, 'Level Sets', bold)
worksheet.write(1, col_gc_rec, 'Grow Cut', bold)
worksheet.write(1, col_ls_rec, 'Level Sets', bold)
# loading data
data_fnames = []
data_slices = []
for name, num in data_struc:
for i in num:
data_fnames.append(name)
data_slices.append(i)
n_imgs = len(data_fnames)
for i, (data_fname, slice) in enumerate(zip(data_fnames, data_slices)):
fname = data_fname.split('/')[-1].split('.')[0]
print '#%i/%i, %s [%i]: ' % (i + 1, n_imgs, fname, slice),
data, gt_mask, voxel_size = tools.load_pickle_data(data_fname)
data = data[slice,...]
gt_mask = gt_mask[slice,...]
print 'windowing ...',
data = tools.windowing(data)
shape_orig = data.shape
data_o = data.copy()
# plt.figure()
# plt.imshow(data, 'gray', vmin=0, vmax=255, interpolation='nearest')
# plt.show()
if scale != 1:
print 'resizing ...',
data = tools.resize_ND(data, scale=scale).astype(np.uint8)
if smoothing:
print 'smoothing ...',
data = tools.smoothing(data)
print 'gc ...',
blob, seg_gc = coarse_segmentation(data, clear_border=False, show=False, verbose=False)
print 'ls ...',
seg_ls = segmentation_refinement(data, blob, show=False, verbose=False)
if seg_gc.shape != gt_mask.shape:
seg_gc = tools.resize_ND(seg_gc, shape=shape_orig)
if seg_ls.shape != gt_mask.shape:
seg_ls = tools.resize_ND(seg_ls, shape=shape_orig)
# computing statistics
print 'stats ...',
precision_gc, recall_gc, f_measure_gc = ac.calc_statistics(blob, gt_mask)
precision_ls, recall_ls, f_measure_ls = ac.calc_statistics(seg_ls, gt_mask)
# writing statistics
print 'writing ...',
items = (f_measure_gc, precision_gc, recall_gc, f_measure_ls, precision_ls, recall_ls)
worksheet.write(start_row + i, 0, fname)
worksheet.write(start_row + i, 1, slice)
for it, col in zip(items, cols_gc + cols_ls):
worksheet.write(start_row + i, col, it)
print 'done'
fig = tools.visualize_seg(data_o, seg_ls, mask=seg_gc, title=fname, show_now=False, for_save=True)
out_fname = '/home/tomas/Dropbox/Data/liver_segmentation/final_alg/%s_sl_%i.png' % (fname, slice)
fig.savefig(out_fname)
plt.close('all')
print '\t\tfmea = %.3f -> %.3f, prec = %.3f -> %.3f, rec = %.3f -> %.3f' %\
(f_measure_gc, f_measure_ls, precision_gc, precision_ls, recall_gc, recall_ls)
else:
mask_res = mask_sfm
stats_sfm = seg_stats(mask, mask_res)
if mask.shape != mask_morph.shape:
mask_res = cv2.resize(mask_morph.astype(np.uint8), mask.shape[::-1])
else:
mask_res = mask_morph
stats_morph = seg_stats(mask, mask_res)
# tp_sfm = 100
# tp_morph = 100
tools.visualize_seg(data,
mask,
mask,
slice=None,
title='ground thruth',
show_now=False)
tools.visualize_seg(
data_sfm,
mask_sfm,
sfm,
slice=None,
title='sfm, TP=%.1f, TN=%.1f, FP=%.1f, FN=%.1f' %
(stats_sfm.tp, stats_sfm.tn, stats_sfm.fp, stats_sfm.fn),
show_now=False)
tools.visualize_seg(
data_morph,
mask_morph,
morph,
slice=None,