def raveler_to_labeled_volume(rav_export_dir, get_glia=False,
use_watershed=True, **kwargs):
"""Import a raveler export stack into a labeled segmented volume."""
import morpho
spmap = read_image_stack(
os.path.join(rav_export_dir, 'superpixel_maps', '*.png'), **kwargs)
sp2seg_list = numpy.loadtxt(
os.path.join(rav_export_dir, 'superpixel_to_segment_map.txt'), uint32)
seg2bod_list = numpy.loadtxt(
os.path.join(rav_export_dir, 'segment_to_body_map.txt'), uint32)
sp2seg = {}
max_sp = sp2seg_list[:,1].max()
start_plane = sp2seg_list[:,0].min()
for z, sp, seg in sp2seg_list:
if not sp2seg.has_key(z):
sp2seg[z] = zeros(max_sp+1, uint32)
sp2seg[z][sp] = seg
max_seg = seg2bod_list[:,0].max()
seg2bod = zeros(max_seg+1, uint32)
seg2bod[seg2bod_list[:,0]] = seg2bod_list[:,1]
initial_output_volume = zeros_like(spmap)
for i, m in enumerate(spmap):
j = start_plane + i
initial_output_volume[i] = seg2bod[sp2seg[j][m]]
probs = kwargs.get('probability_map', ones_like(spmap))
output_volume = morpho.watershed(probs, seeds=initial_output_volume) \
if use_watershed else initial_output_volume
if get_glia:
annots = json.load(
open(os.path.join(rav_export_dir, 'annotations-body.json'), 'r'))
glia = [a['body ID'] for a in annots['data']
if a.get('comment', None) == 'glia']
return output_volume, glia
else:
return output_volume
def raveler_to_labeled_volume(rav_export_dir,
get_glia=False,
use_watershed=False,
**kwargs):
"""Import a raveler export stack into a labeled segmented volume."""
import morpho
spmap = read_image_stack(
os.path.join(rav_export_dir, 'superpixel_maps', '*.png'), **kwargs)
sp2seg_list = numpy.loadtxt(
os.path.join(rav_export_dir, 'superpixel_to_segment_map.txt'), uint32)
seg2bod_list = numpy.loadtxt(
os.path.join(rav_export_dir, 'segment_to_body_map.txt'), uint32)
sp2seg = {}
max_sp = sp2seg_list[:, 1].max()
start_plane = sp2seg_list[:, 0].min()
for z, sp, seg in sp2seg_list:
if not sp2seg.has_key(z):
sp2seg[z] = zeros(max_sp + 1, uint32)
sp2seg[z][sp] = seg
max_seg = seg2bod_list[:, 0].max()
seg2bod = zeros(max_seg + 1, uint32)
seg2bod[seg2bod_list[:, 0]] = seg2bod_list[:, 1]
initial_output_volume = zeros_like(spmap)
for i, m in enumerate(spmap):
j = start_plane + i
initial_output_volume[i] = seg2bod[sp2seg[j][m]]
probs = kwargs.get('probability_map', ones_like(spmap))
output_volume = morpho.watershed(probs, seeds=initial_output_volume) \
if use_watershed else initial_output_volume
if (output_volume[:, 0, 0] == 0).all() and \
(output_volume == 0).sum() == output_volume.shape[0]:
output_volume[:, 0, 0] = output_volume[:, 0, 1]
if get_glia:
annots = json.load(
open(os.path.join(rav_export_dir, 'annotations-body.json'), 'r'))
glia = [
a['body ID'] for a in annots['data']
if a.get('comment', None) == 'glia'
]
return output_volume, glia
else:
return output_volume
def raveler_to_labeled_volume(rav_export_dir, get_glia=False,
use_watershed=False, probability_map=None, crop=None):
"""Import a raveler export stack into a labeled segmented volume.
Parameters
----------
rav_export_dir : string
The directory containing the Raveler stack.
get_glia : bool (optional, default False)
Return the segment numbers corresponding to glia, if available.
use_watershed : bool (optional, default False)
Fill in 0-labeled voxels using watershed.
probability_map : np.ndarray, same shape as volume to be read (optional)
If `use_watershed` is True, use `probability_map` as the landscape. If
this is not provided, it uses a flat landscape.
crop : tuple of int (optional, default None)
A 6-tuple of [xmin, xmax, ymin, ymax, zmin, zmax].
Returns
-------
output_volume : np.ndarray, shape (Z, X, Y)
The segmentation in the Raveler volume.
glia : list of int (optional, only returned if `get_glia` is True)
The IDs in the segmentation corresponding to glial cells.
"""
import morpho
spmap = read_image_stack(
os.path.join(rav_export_dir, 'superpixel_maps', '*.png'), crop=crop)
spmap = raveler_rgba_to_int(spmap)
sp2seg_list = np.loadtxt(
os.path.join(rav_export_dir, 'superpixel_to_segment_map.txt'), uint32)
seg2bod_list = np.loadtxt(
os.path.join(rav_export_dir, 'segment_to_body_map.txt'), uint32)
sp2seg = {}
max_sp = sp2seg_list[:,1].max()
start_plane = sp2seg_list[:,0].min()
for z, sp, seg in sp2seg_list:
if not sp2seg.has_key(z):
sp2seg[z] = zeros(max_sp+1, uint32)
sp2seg[z][sp] = seg
max_seg = seg2bod_list[:,0].max()
seg2bod = zeros(max_seg+1, uint32)
seg2bod[seg2bod_list[:,0]] = seg2bod_list[:,1]
initial_output_volume = zeros_like(spmap)
for i, m in enumerate(spmap):
j = start_plane + i
initial_output_volume[i] = seg2bod[sp2seg[j][m]]
if use_watershed:
probs = np.ones_like(spmap) if probability_map is None \
else probability_map
output_volume = morpho.watershed(probs, seeds=initial_output_volume)
else:
output_volume = initial_output_volume
if (output_volume[:, 0, 0] == 0).all() and \
(output_volume == 0).sum() == output_volume.shape[0]:
output_volume[:, 0, 0] = output_volume[:, 0, 1]
if get_glia:
annots = json.load(
open(os.path.join(rav_export_dir, 'annotations-body.json'), 'r'))
glia = [a['body ID'] for a in annots['data']
if a.get('comment', None) == 'glia']
return output_volume, glia
else:
return output_volume
def raveler_to_labeled_volume(rav_export_dir,
get_glia=False,
use_watershed=False,
probability_map=None,
crop=None):
"""Import a raveler export stack into a labeled segmented volume.
Parameters
----------
rav_export_dir : string
The directory containing the Raveler stack.
get_glia : bool (optional, default False)
Return the segment numbers corresponding to glia, if available.
use_watershed : bool (optional, default False)
Fill in 0-labeled voxels using watershed.
probability_map : np.ndarray, same shape as volume to be read (optional)
If `use_watershed` is True, use `probability_map` as the landscape. If
this is not provided, it uses a flat landscape.
crop : tuple of int (optional, default None)
A 6-tuple of [xmin, xmax, ymin, ymax, zmin, zmax].
Returns
-------
output_volume : np.ndarray, shape (Z, X, Y)
The segmentation in the Raveler volume.
glia : list of int (optional, only returned if `get_glia` is True)
The IDs in the segmentation corresponding to glial cells.
"""
import morpho
spmap = read_image_stack(os.path.join(rav_export_dir, 'superpixel_maps',
'*.png'),
crop=crop)
spmap = raveler_rgba_to_int(spmap)
sp2seg_list = np.loadtxt(
os.path.join(rav_export_dir, 'superpixel_to_segment_map.txt'), uint32)
seg2bod_list = np.loadtxt(
os.path.join(rav_export_dir, 'segment_to_body_map.txt'), uint32)
sp2seg = {}
max_sp = sp2seg_list[:, 1].max()
start_plane = sp2seg_list[:, 0].min()
for z, sp, seg in sp2seg_list:
if not sp2seg.has_key(z):
sp2seg[z] = zeros(max_sp + 1, uint32)
sp2seg[z][sp] = seg
max_seg = seg2bod_list[:, 0].max()
seg2bod = zeros(max_seg + 1, uint32)
seg2bod[seg2bod_list[:, 0]] = seg2bod_list[:, 1]
initial_output_volume = zeros_like(spmap)
for i, m in enumerate(spmap):
j = start_plane + i
initial_output_volume[i] = seg2bod[sp2seg[j][m]]
if use_watershed:
probs = np.ones_like(spmap) if probability_map is None \
else probability_map
output_volume = morpho.watershed(probs, seeds=initial_output_volume)
else:
output_volume = initial_output_volume
if (output_volume[:, 0, 0] == 0).all() and \
(output_volume == 0).sum() == output_volume.shape[0]:
output_volume[:, 0, 0] = output_volume[:, 0, 1]
if get_glia:
annots = json.load(
open(os.path.join(rav_export_dir, 'annotations-body.json'), 'r'))
glia = [
a['body ID'] for a in annots['data']
if a.get('comment', None) == 'glia'
]
return output_volume, glia
else:
return output_volume