def prepare_chaincode_features(data, dictionary_name=None, bow_options_name=None, input_name=None, output_dir=None, **kwargs):
assert bow_options_name != None
assert dictionary_name != None
assert input_name != None
assert output_dir != None
bow_options = tsh.deserialize(bow_options_name)
dictionary_name = dictionary_name.format(OUT=output_dir)
ret = { 'bow_options': bow_options, 'dictionary_name': dictionary_name }
if os.path.exists(dictionary_name):
ret.update(tsh.deserialize(dictionary_name))
ret['feature_names'] = [ 's%02di%03d' % (scale, i) for scale in bow_options['scales'] for i in range(len(ret['dictionary'][scale])) ]
return ret
# Compute dictionary - for this we need the chaincodes
chaincode_dtype = [ ('id', data.dtype['id']) ] + zip(
[ 'cc%02d' % scale for scale in bow_options['scales'] ],
['O'] * len(bow_options['scales']))
chaincodes = np.zeros(len(data), dtype=chaincode_dtype)
chaincodes['id'] = data['id']
mask_prefix = os.path.expanduser(kwargs['mask_prefix'])
for scale in bow_options['scales']:
scale_name = 'cc%02d' % scale
for i in range(len(data)):
chaincodes[i][scale_name] = get_chaincode_from_image(os.path.join(mask_prefix, data[i]['mask']), scale)
tsh.serialize(os.path.join(output_dir, input_name + '-chaincodes.dat'), chaincodes)
lpname = os.path.join(output_dir, input_name + '-lp.dat')
dictionary = dict(zip(
['dictionary', 'counts'],
train_dictionary(chaincodes, lpname=lpname, labels=data[kwargs['truth']], **bow_options)))
tsh.serialize(dictionary_name, dictionary)
ret.update(dictionary)
ret['feature_names'] = [ 's%02di%03d' % (scale, i) for scale in bow_options['scales'] for i in range(len(ret['dictionary'][scale])) ]
return ret
def prepare_chaincode_features(data,
dictionary_name=None,
bow_options_name=None,
input_name=None,
output_dir=None,
**kwargs):
assert bow_options_name != None
assert dictionary_name != None
assert input_name != None
assert output_dir != None
bow_options = tsh.deserialize(bow_options_name)
dictionary_name = dictionary_name.format(OUT=output_dir)
ret = {'bow_options': bow_options, 'dictionary_name': dictionary_name}
if os.path.exists(dictionary_name):
ret.update(tsh.deserialize(dictionary_name))
ret['feature_names'] = [
's%02di%03d' % (scale, i) for scale in bow_options['scales']
for i in range(len(ret['dictionary'][scale]))
]
return ret
# Compute dictionary - for this we need the chaincodes
chaincode_dtype = [
('id', data.dtype['id'])
] + zip(['cc%02d' % scale for scale in bow_options['scales']],
['O'] * len(bow_options['scales']))
chaincodes = np.zeros(len(data), dtype=chaincode_dtype)
chaincodes['id'] = data['id']
mask_prefix = os.path.expanduser(kwargs['mask_prefix'])
for scale in bow_options['scales']:
scale_name = 'cc%02d' % scale
for i in range(len(data)):
chaincodes[i][scale_name] = get_chaincode_from_image(
os.path.join(mask_prefix, data[i]['mask']), scale)
tsh.serialize(os.path.join(output_dir, input_name + '-chaincodes.dat'),
chaincodes)
lpname = os.path.join(output_dir, input_name + '-lp.dat')
dictionary = dict(
zip(['dictionary', 'counts'],
train_dictionary(chaincodes,
lpname=lpname,
labels=data[kwargs['truth']],
**bow_options)))
tsh.serialize(dictionary_name, dictionary)
ret.update(dictionary)
ret['feature_names'] = [
's%02di%03d' % (scale, i) for scale in bow_options['scales']
for i in range(len(ret['dictionary'][scale]))
]
return ret
def read_argsfile(filename):
"""
Reads in arguments from file.
Parameters:
-----------
filename: string
Input filename.
"""
return tsh.deserialize(filename)
def extract_expression(image_file, mask_file, inside_file, expression_file,
data_file, normalized_width, normalized_height):
'''
'''
if os.path.exists(data_file):
return tsh.deserialize(data_file)
image = tsh.read_gray_image(image_file)
mask = tsh.read_gray_image(mask_file)
inside, expression = tsh.extract_embryo_gray(image, mask > 0, normalized_width, normalized_height)
if inside_file is not None:
cv.imwrite(inside_file, inside)
if expression_file is not None:
cv.imwrite(expression_file, 255*expression)
tsh.serialize(data_file, expression)
return expression
def extract_expression(image_file, mask_file, inside_file, expression_file,
data_file, normalized_width, normalized_height):
'''
'''
if os.path.exists(data_file):
return tsh.deserialize(data_file)
image = tsh.read_gray_image(image_file)
mask = tsh.read_gray_image(mask_file)
inside, expression = tsh.extract_embryo_gray(image, mask > 0,
normalized_width,
normalized_height)
if inside_file is not None:
cv.imwrite(inside_file, inside)
if expression_file is not None:
cv.imwrite(expression_file, 255 * expression)
tsh.serialize(data_file, expression)
return expression
def is_blurred(fine_segmentation, dirname, ids):
data = tsh.deserialize(os.path.join(fine_segmentation, dirname, 'blur-evaluation.dat'))
return [ data['data'][data['data']['obj'] == int(metasys_id)][0][data['blur_measure']] < 1.46128831851
for metasys_id in ids ]
def _get_dissimilarity(i, j, measure, rotation_invariance, data_filei,
data_filej):
expri = tsh.deserialize(data_filei).astype(float)
exprj = tsh.deserialize(data_filej).astype(float)
d, t = image_distance(expri, exprj, measure, rotation_invariance)
return i, j, d, t
def get_dissimilarities(data,
output_dir=None,
input_name=None,
image_prefix=None,
mask_prefix=None,
n_jobs=None,
**kwargs):
assert image_prefix != None
assert mask_prefix != None
assert output_dir != None
assert input_name != None
measure = kwargs['measure']
distance_name = kwargs['distance_name']
rotation_invariance = kwargs['rotation_invariance']
normalized_width = kwargs['normalized_width']
normalized_height = kwargs['normalized_height']
if n_jobs == None:
n_jobs = 1
image_prefix = os.path.expanduser(image_prefix)
mask_prefix = os.path.expanduser(image_prefix)
distance_name = distance_name.format(OUT=output_dir,
INPUTNAME=input_name,
**kwargs)
tsh.makedirs(os.path.dirname(distance_name))
kwargs['distance_name'] = distance_name
expr_dir = os.path.join(
output_dir, 'expr/%04dx%04d' % (normalized_width, normalized_height))
tsh.makedirs(expr_dir)
# Make it easier for evaluate.py to create nice html reports.
if 'create_links' in kwargs and kwargs['create_links'] == True:
if os.path.exists('expr'):
os.unlink('expr')
try:
os.symlink(expr_dir, 'expr')
except:
pass
if os.path.exists('distance'):
os.unlink('distance')
try:
os.symlink(os.path.dirname(distance_name), 'distance')
except:
pass
save_expr_images = kwargs[
'save_expr_images'] if 'save_expr_images' in kwargs else False
if os.path.exists(distance_name):
D = tsh.deserialize(distance_name)['D']
else:
imagenames = [
os.path.join(image_prefix, sample['image']) for sample in data
]
masknames = [
os.path.join(mask_prefix, sample['mask']) for sample in data
]
n = len(data)
logger.info('Extracting %d expressions...', n)
Parallel(n_jobs=n_jobs, verbose=True, pre_dispatch='2*n_jobs')(
delayed(_extract_expression)
(imagenames[j], masknames[j],
os.path.join(expr_dir, 'inside%02d.png' %
data[j]['id']) if save_expr_images else None,
os.path.join(expr_dir, 'expr%02d.png' %
data[j]['id']) if save_expr_images else None,
os.path.join(expr_dir, 'expr%02d.dat' %
data[j]['id']), normalized_width, normalized_height)
for j in range(n))
logger.info('Computing %d dissimilarities...', (n * (n - 1)) / 2)
results = Parallel(
n_jobs=n_jobs, verbose=True,
pre_dispatch='2*n_jobs')(delayed(_get_dissimilarity)(
i, j, measure, rotation_invariance,
os.path.join(expr_dir, 'expr%02d.dat' % data[i]['id']),
os.path.join(expr_dir, 'expr%02d.dat' % data[j]['id']))
for j in range(n)
for i in range(j + 1, n))
logger.info('Transforming results...')
D = np.zeros((n, n), dtype=np.float)
tfxs = np.array([['I'] * n] * n)
for i, j, d, t in results:
D[j, i] = d
D[i, j] = d
tfxs[j, i] = t
tfxs[i, j] = t
logger.info('Saving results...')
tsh.serialize(
distance_name, {
'D': D,
'min': None,
'max': None,
'tfxs': tfxs,
'measure': measure,
'rotation_invariance': rotation_invariance
})
return kwargs, D
def _get_dissimilarity(i, j, measure, rotation_invariance, data_filei, data_filej):
expri = tsh.deserialize(data_filei).astype(float)
exprj = tsh.deserialize(data_filej).astype(float)
d, t = image_distance(expri, exprj, measure, rotation_invariance)
return i, j, d, t
def get_dissimilarities(data, output_dir=None, input_name=None, image_prefix=None, mask_prefix=None, n_jobs=None, **kwargs):
assert image_prefix != None
assert mask_prefix != None
assert output_dir != None
assert input_name != None
measure = kwargs['measure']
distance_name = kwargs['distance_name']
rotation_invariance = kwargs['rotation_invariance']
normalized_width = kwargs['normalized_width']
normalized_height = kwargs['normalized_height']
if n_jobs == None:
n_jobs = 1
image_prefix = os.path.expanduser(image_prefix)
mask_prefix = os.path.expanduser(image_prefix)
distance_name = distance_name.format(OUT=output_dir, INPUTNAME=input_name, **kwargs)
tsh.makedirs(os.path.dirname(distance_name))
kwargs['distance_name'] = distance_name
expr_dir = os.path.join(output_dir, 'expr/%04dx%04d' % (normalized_width, normalized_height))
tsh.makedirs(expr_dir)
# Make it easier for evaluate.py to create nice html reports.
if 'create_links' in kwargs and kwargs['create_links'] == True:
if os.path.exists('expr'):
os.unlink('expr')
try:
os.symlink(expr_dir, 'expr')
except:
pass
if os.path.exists('distance'):
os.unlink('distance')
try:
os.symlink(os.path.dirname(distance_name), 'distance')
except:
pass
save_expr_images = kwargs['save_expr_images'] if 'save_expr_images' in kwargs else False
if os.path.exists(distance_name):
D = tsh.deserialize(distance_name)['D']
else:
imagenames = [ os.path.join(image_prefix, sample['image']) for sample in data ]
masknames = [ os.path.join(mask_prefix, sample['mask']) for sample in data ]
n = len(data)
logger.info('Extracting %d expressions...', n)
Parallel(n_jobs=n_jobs, verbose=True,
pre_dispatch='2*n_jobs')(
delayed(_extract_expression)(
imagenames[j],
masknames[j],
os.path.join(expr_dir, 'inside%02d.png' % data[j]['id']) if save_expr_images else None,
os.path.join(expr_dir, 'expr%02d.png' % data[j]['id']) if save_expr_images else None,
os.path.join(expr_dir, 'expr%02d.dat' % data[j]['id']),
normalized_width,
normalized_height
) for j in range(n))
logger.info('Computing %d dissimilarities...', (n*(n-1))/2)
results = Parallel(n_jobs=n_jobs, verbose=True,
pre_dispatch='2*n_jobs')(
delayed(_get_dissimilarity)(
i, j,
measure, rotation_invariance,
os.path.join(expr_dir, 'expr%02d.dat' % data[i]['id']),
os.path.join(expr_dir, 'expr%02d.dat' % data[j]['id'])
) for j in range(n) for i in range(j+1, n))
logger.info('Transforming results...')
D = np.zeros((n, n), dtype=np.float)
tfxs = np.array([['I'] * n] * n)
for i, j, d, t in results:
D[j, i] = d
D[i, j] = d
tfxs[j, i] = t
tfxs[i, j] = t
logger.info('Saving results...')
tsh.serialize(distance_name, {
'D': D,
'min': None, 'max': None,
'tfxs': tfxs,
'measure': measure,
'rotation_invariance': rotation_invariance })
return kwargs, D
def read_propagatorfile(filename):
return tsh.deserialize(filename)
def read_classifierfile(filename):
return tsh.deserialize(filename)