class LineProfiler(Viewer):
"""LineProfiler widget class."""
_view_name = Unicode('LineProfilerView').tag(sync=True)
_model_name = Unicode('LineProfilerModel').tag(sync=True)
_view_module = Unicode('itk-jupyter-widgets').tag(sync=True)
_model_module = Unicode('itk-jupyter-widgets').tag(sync=True)
_view_module_version = Unicode('^0.17.0').tag(sync=True)
_model_module_version = Unicode('^0.17.0').tag(sync=True)
point1 = NDArray(dtype=np.float64, default_value=np.zeros((3,), dtype=np.float64),
help="First point in physical space that defines the line profile")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(3,))
point2 = NDArray(dtype=np.float64, default_value=np.ones((3,), dtype=np.float64),
help="First point in physical space that defines the line profile")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(3,))
_select_initial_points = CBool(
default_value=False,
help="We will select the initial points for the line profile.").tag(
sync=True)
def __init__(self, **kwargs):
if 'point1' not in kwargs or 'point2' not in kwargs:
self._select_initial_points = True
# Default to z-plane mode instead of the 3D volume if we need to
# select points
if 'mode' not in kwargs:
kwargs['mode'] = 'z'
if 'ui_collapsed' not in kwargs:
kwargs['ui_collapsed'] = True
super(LineProfiler, self).__init__(**kwargs)
class Viewer3d(DOMWidget):
_model_name = Unicode('Viewer3dModel').tag(sync=True)
_model_module = Unicode(MODULE_NAME).tag(sync=True)
_model_module_version = Unicode(EXTENSION_SPEC_VERSION).tag(sync=True)
_view_name = Unicode('Viewer3dView').tag(sync=True)
_view_module = Unicode(MODULE_NAME).tag(sync=True)
_view_module_version = Unicode(EXTENSION_SPEC_VERSION).tag(sync=True)
image = NDArray(dtype=np.float64,
default_value=np.zeros(0, dtype=np.float64)).tag(
sync=True, **array_serialization)
# server = Unicode("http://atlantis.sci.utah.edu").tag(sync=True)
# dataset = Unicode('').tag(sync=True)
# tile_size = Int(512).tag(sync=True)
def __init__(self, *args, **kwargs):
super().__init__(**kwargs)
class Viewer(ViewerParent):
"""Viewer widget class."""
_view_name = Unicode('ViewerView').tag(sync=True)
_model_name = Unicode('ViewerModel').tag(sync=True)
_view_module = Unicode('itkwidgets').tag(sync=True)
_model_module = Unicode('itkwidgets').tag(sync=True)
_view_module_version = Unicode('^0.26.1').tag(sync=True)
_model_module_version = Unicode('^0.26.1').tag(sync=True)
image = ITKImage(
default_value=None,
allow_none=True,
help="Image to visualize.").tag(
sync=False,
**itkimage_serialization)
rendered_image = ITKImage(
default_value=None,
allow_none=True).tag(
sync=True,
**itkimage_serialization)
_rendering_image = CBool(
default_value=False,
help="We are currently volume rendering the image.").tag(sync=True)
interpolation = CBool(
default_value=True,
help="Use linear interpolation in slicing planes.").tag(sync=True)
cmap = Colormap('Viridis (matplotlib)').tag(sync=True)
_custom_cmap = NDArray(dtype=np.float32, default_value=None, allow_none=True,
help="RGB triples from 0.0 to 1.0 that define a custom linear, sequential colormap")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(None, 3))
shadow = CBool(
default_value=True,
help="Use shadowing in the volume rendering.").tag(sync=True)
slicing_planes = CBool(
default_value=False,
help="Display the slicing planes in volume rendering view mode.").tag(
sync=True)
x_slice = CFloat(
default_value=None,
allow_none=True,
help="World-space position of the X slicing plane.").tag(sync=True)
y_slice = CFloat(
default_value=None,
allow_none=True,
help="World-space position of the Y slicing plane.").tag(sync=True)
z_slice = CFloat(
default_value=None,
allow_none=True,
help="World-space position of the Z slicing plane.").tag(sync=True)
gradient_opacity = CFloat(
default_value=0.2,
help="Volume rendering gradient opacity, from (0.0, 1.0]").tag(sync=True)
blend = CaselessStrEnum(
('composite',
'max',
'min',
'average'),
default_value='composite',
help="Volume rendering blend mode").tag(sync=True)
roi = NDArray(dtype=np.float64, default_value=np.zeros((2, 3), dtype=np.float64),
help="Region of interest: [[lower_x, lower_y, lower_z), (upper_x, upper_y, upper_z]]")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(2, 3))
vmin = CFloat(
default_value=None,
allow_none=True,
help="Value that maps to the minimum of image colormap.").tag(
sync=True)
vmax = CFloat(
default_value=None,
allow_none=True,
help="Value that maps to the maximum of image colormap.").tag(
sync=True)
_largest_roi = NDArray(dtype=np.float64, default_value=np.zeros((2, 3), dtype=np.float64),
help="Largest possible region of interest: "
"[[lower_x, lower_y, lower_z), (upper_x, upper_y, upper_z]]")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(2, 3))
select_roi = CBool(
default_value=False,
help="Enable an interactive region of interest widget for the image.").tag(
sync=True)
size_limit_2d = NDArray(dtype=np.int64, default_value=np.array([1024, 1024], dtype=np.int64),
help="Size limit for 2D image visualization.").tag(sync=False)
size_limit_3d = NDArray(dtype=np.int64, default_value=np.array([192, 192, 192], dtype=np.int64),
help="Size limit for 3D image visualization.").tag(sync=False)
_scale_factors = NDArray(dtype=np.uint8, default_value=np.array([1, 1, 1], dtype=np.uint8),
help="Image downscaling factors.").tag(sync=True, **array_serialization)
_downsampling = CBool(default_value=False,
help="We are downsampling the image to meet the size limits.").tag(sync=True)
_reset_crop_requested = CBool(default_value=False,
help="The user requested a reset of the roi.").tag(sync=True)
units = Unicode(
'',
help="Units to display in the scale bar.").tag(
sync=True)
point_set_representations = List(
trait=Unicode(),
default_value=[],
help="Point set representation").tag(
sync=True)
point_sets = PointSetList(
default_value=None,
allow_none=True,
help="Point sets to visualize").tag(
sync=True,
**polydata_list_serialization)
point_set_colors = NDArray(dtype=np.float32, default_value=np.zeros((0, 3), dtype=np.float32),
help="RGB colors for the points sets")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(None, 3))
point_set_opacities = NDArray(dtype=np.float32, default_value=np.zeros((0,), dtype=np.float32),
help="Opacities for the points sets")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(None,))
point_set_representations = List(
trait=Unicode(),
default_value=[],
help="Point set representation").tag(
sync=True)
geometries = PolyDataList(
default_value=None,
allow_none=True,
help="Geometries to visualize").tag(
sync=True,
**polydata_list_serialization)
geometry_colors = NDArray(dtype=np.float32, default_value=np.zeros((0, 3), dtype=np.float32),
help="RGB colors for the geometries")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(None, 3))
geometry_opacities = NDArray(dtype=np.float32, default_value=np.zeros((0,), dtype=np.float32),
help="Opacities for the geometries")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(None,))
ui_collapsed = CBool(
default_value=False,
help="Collapse the built in user interface.").tag(
sync=True)
rotate = CBool(
default_value=False,
help="Rotate the camera around the scene.").tag(
sync=True)
annotations = CBool(
default_value=True,
help="Show annotations.").tag(
sync=True)
mode = CaselessStrEnum(
('x',
'y',
'z',
'v'),
default_value='v',
help="View mode: x: x plane, y: y plane, z: z plane, v: volume rendering").tag(
sync=True)
camera = NDArray(dtype=np.float32, default_value=np.zeros((3, 3), dtype=np.float32),
help="Camera parameters: [[position_x, position_y, position_z], "
"[focal_point_x, focal_point_y, focal_point_z], "
"[view_up_x, view_up_y, view_up_z]]")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(3, 3))
def __init__(self, **kwargs): # noqa: C901
if 'point_set_colors' in kwargs:
proposal = {'value': kwargs['point_set_colors']}
color_array = self._validate_point_set_colors(proposal)
kwargs['point_set_colors'] = color_array
if 'point_set_opacities' in kwargs:
proposal = {'value': kwargs['point_set_opacities']}
opacities_array = self._validate_point_set_opacities(proposal)
kwargs['point_set_opacities'] = opacities_array
if 'point_set_representations' in kwargs:
proposal = {'value': kwargs['point_set_representations']}
representations_list = self._validate_point_set_representations(
proposal)
kwargs['point_set_representations'] = representations_list
self.observe(self._on_point_sets_changed, ['point_sets'])
if 'geometry_colors' in kwargs:
proposal = {'value': kwargs['geometry_colors']}
color_array = self._validate_geometry_colors(proposal)
kwargs['geometry_colors'] = color_array
if 'geometry_opacities' in kwargs:
proposal = {'value': kwargs['geometry_opacities']}
opacities_array = self._validate_geometry_opacities(proposal)
kwargs['geometry_opacities'] = opacities_array
self.observe(self._on_geometries_changed, ['geometries'])
super(Viewer, self).__init__(**kwargs)
if not self.image:
return
dimension = self.image.GetImageDimension()
largest_region = self.image.GetLargestPossibleRegion()
size = largest_region.GetSize()
# Cache this so we do not need to recompute on it when resetting the
# roi
self._largest_roi_rendered_image = None
self._largest_roi = np.zeros((2, 3), dtype=np.float64)
if not np.any(self.roi):
largest_index = largest_region.GetIndex()
self.roi[0][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index))
largest_index_upper = largest_index + size
self.roi[1][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index_upper))
self._largest_roi = self.roi.copy()
if dimension == 2:
for dim in range(dimension):
if size[dim] > self.size_limit_2d[dim]:
self._downsampling = True
else:
for dim in range(dimension):
if size[dim] > self.size_limit_3d[dim]:
self._downsampling = True
if self._downsampling:
self.extractor = itk.ExtractImageFilter.New(self.image)
self.shrinker = itk.BinShrinkImageFilter.New(self.extractor)
self._update_rendered_image()
if self._downsampling:
self.observe(self._on_roi_changed, ['roi'])
self.observe(self._on_reset_crop_requested, ['_reset_crop_requested'])
self.observe(self.update_rendered_image, ['image'])
def _on_roi_changed(self, change=None):
if self._downsampling:
self._update_rendered_image()
def _on_reset_crop_requested(self, change=None):
if change.new is True and self._downsampling:
dimension = self.image.GetImageDimension()
largest_region = self.image.GetLargestPossibleRegion()
size = largest_region.GetSize()
largest_index = largest_region.GetIndex()
new_roi = self.roi.copy()
new_roi[0][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index))
largest_index_upper = largest_index + size
new_roi[1][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index_upper))
self._largest_roi = new_roi.copy()
self.roi = new_roi
if change.new is True:
self._reset_crop_requested = False
@debounced(delay_seconds=0.2, method=True)
def update_rendered_image(self, change=None):
self._largest_roi_rendered_image = None
self._largest_roi = np.zeros((2, 3), dtype=np.float64)
self._update_rendered_image()
@staticmethod
def _find_scale_factors(limit, dimension, size):
scale_factors = [1, ] * 3
for dim in range(dimension):
while(int(np.floor(float(size[dim]) / scale_factors[dim])) > limit[dim]):
scale_factors[dim] += 1
return scale_factors
def _update_rendered_image(self):
if self.image is None:
return
if self._rendering_image:
@yield_for_change(self, '_rendering_image')
def f():
x = yield
assert(x is False)
f()
self._rendering_image = True
if self._downsampling:
dimension = self.image.GetImageDimension()
index = self.image.TransformPhysicalPointToIndex(
self.roi[0][:dimension])
upper_index = self.image.TransformPhysicalPointToIndex(
self.roi[1][:dimension])
size = upper_index - index
if dimension == 2:
scale_factors = self._find_scale_factors(
self.size_limit_2d, dimension, size)
else:
scale_factors = self._find_scale_factors(
self.size_limit_3d, dimension, size)
self._scale_factors = np.array(scale_factors, dtype=np.uint8)
self.shrinker.SetShrinkFactors(scale_factors[:dimension])
region = itk.ImageRegion[dimension]()
region.SetIndex(index)
region.SetSize(tuple(size))
# Account for rounding
# truncation issues
region.PadByRadius(1)
region.Crop(self.image.GetLargestPossibleRegion())
self.extractor.SetInput(self.image)
self.extractor.SetExtractionRegion(region)
size = region.GetSize()
is_largest = False
if np.any(self._largest_roi) and np.all(
self._largest_roi == self.roi):
is_largest = True
if self._largest_roi_rendered_image is not None:
self.rendered_image = self._largest_roi_rendered_image
return
self.shrinker.UpdateLargestPossibleRegion()
if is_largest:
self._largest_roi_rendered_image = self.shrinker.GetOutput()
self._largest_roi_rendered_image.DisconnectPipeline()
self._largest_roi_rendered_image.SetOrigin(
self.roi[0][:dimension])
self.rendered_image = self._largest_roi_rendered_image
return
shrunk = self.shrinker.GetOutput()
shrunk.DisconnectPipeline()
shrunk.SetOrigin(self.roi[0][:dimension])
self.rendered_image = shrunk
else:
self.rendered_image = self.image
@validate('gradient_opacity')
def _validate_gradient_opacity(self, proposal):
"""Enforce 0 < value <= 1.0."""
value = proposal['value']
if value <= 0.0:
return 0.01
if value > 1.0:
return 1.0
return value
@validate('point_set_colors')
def _validate_point_set_colors(self, proposal):
value = proposal['value']
n_colors = len(value)
if self.point_sets:
n_colors = len(self.point_sets)
result = np.zeros((n_colors, 3), dtype=np.float32)
for index, color in enumerate(value):
result[index, :] = matplotlib.colors.to_rgb(color)
if len(value) < n_colors:
for index in range(len(value), n_colors):
color = colorcet.glasbey[index % len(colorcet.glasbey)]
result[index, :] = matplotlib.colors.to_rgb(color)
return result
@validate('point_set_opacities')
def _validate_point_set_opacities(self, proposal):
value = proposal['value']
n_values = 0
if isinstance(value, float):
n_values = 1
else:
n_values = len(value)
n_opacities = n_values
if self.point_sets:
n_opacities = len(self.point_sets)
result = np.ones((n_opacities,), dtype=np.float32)
result[:n_values] = value
return result
@validate('point_set_representations')
def _validate_point_set_representations(self, proposal):
value = proposal['value']
n_values = 0
if isinstance(value, str):
n_values = 1
else:
n_values = len(value)
n_representations = n_values
if self.point_sets:
n_representations = len(self.point_sets)
result = ['points'] * n_representations
result[:n_values] = value
return result
def _on_point_sets_changed(self, change=None):
# Make sure we have a sufficient number of colors
old_colors = self.point_set_colors
self.point_set_colors = old_colors[:len(self.point_sets)]
# Make sure we have a sufficient number of opacities
old_opacities = self.point_set_opacities
self.point_set_opacities = old_opacities[:len(self.point_sets)]
# Make sure we have a sufficient number of representations
old_representations = self.point_set_representations
self.point_set_representations = old_representations[:len(
self.point_sets)]
@validate('geometry_colors')
def _validate_geometry_colors(self, proposal):
value = proposal['value']
n_colors = len(value)
if self.geometries:
n_colors = len(self.geometries)
result = np.zeros((n_colors, 3), dtype=np.float32)
for index, color in enumerate(value):
result[index, :] = matplotlib.colors.to_rgb(color)
if len(value) < n_colors:
for index in range(len(value), n_colors):
color = colorcet.glasbey[index % len(colorcet.glasbey)]
result[index, :] = matplotlib.colors.to_rgb(color)
return result
@validate('geometry_opacities')
def _validate_geometry_opacities(self, proposal):
value = proposal['value']
n_values = 0
if isinstance(value, float):
n_values = 1
else:
n_values = len(value)
n_opacities = n_values
if self.geometries:
n_opacities = len(self.geometries)
result = np.ones((n_opacities,), dtype=np.float32)
result[:n_values] = value
return result
def _on_geometries_changed(self, change=None):
# Make sure we have a sufficient number of colors
old_colors = self.geometry_colors
self.geometry_colors = old_colors[:len(self.geometries)]
# Make sure we have a sufficient number of opacities
old_opacities = self.geometry_opacities
self.geometry_opacities = old_opacities[:len(self.geometries)]
def roi_region(self):
"""Return the itk.ImageRegion corresponding to the roi."""
dimension = self.image.GetImageDimension()
index = self.image.TransformPhysicalPointToIndex(
tuple(self.roi[0][:dimension]))
upper_index = self.image.TransformPhysicalPointToIndex(
tuple(self.roi[1][:dimension]))
size = upper_index - index
for dim in range(dimension):
size[dim] += 1
region = itk.ImageRegion[dimension]()
region.SetIndex(index)
region.SetSize(tuple(size))
region.Crop(self.image.GetLargestPossibleRegion())
return region
def roi_slice(self):
"""Return the numpy array slice corresponding to the roi."""
dimension = self.image.GetImageDimension()
region = self.roi_region()
index = region.GetIndex()
upper_index = np.array(index) + np.array(region.GetSize())
slices = []
for dim in range(dimension):
slices.insert(0, slice(index[dim], upper_index[dim] + 1))
return tuple(slices)
class LineProfiler(Viewer):
"""LineProfiler widget class."""
_view_name = Unicode('LineProfilerView').tag(sync=True)
_model_name = Unicode('LineProfilerModel').tag(sync=True)
_view_module = Unicode('itkwidgets').tag(sync=True)
_model_module = Unicode('itkwidgets').tag(sync=True)
_view_module_version = Unicode('^0.32.0').tag(sync=True)
_model_module_version = Unicode('^0.32.0').tag(sync=True)
point1 = NDArray(dtype=np.float64, default_value=np.zeros((3,), dtype=np.float64),
help="First point in physical space that defines the line profile")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(3,))
point2 = NDArray(dtype=np.float64, default_value=np.ones((3,), dtype=np.float64),
help="First point in physical space that defines the line profile")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(3,))
_select_initial_points = CBool(
default_value=False,
help="We will select the initial points for the line profile.").tag(
sync=True)
def __init__(self, image, order, **kwargs):
self.image = image
self.order = order
if 'point1' not in kwargs or 'point2' not in kwargs:
self._select_initial_points = True
# Default to z-plane mode instead of the 3D volume if we need to
# select points
if 'mode' not in kwargs:
kwargs['mode'] = 'z'
if 'ui_collapsed' not in kwargs:
kwargs['ui_collapsed'] = True
super(LineProfiler, self).__init__(**kwargs)
def get_profile(self,
image_or_array=None,
point1=None,
point2=None,
order=None):
"""Calculate the line profile.
Calculate the pixel intensity values along the line that connects
the given two points.
The image can be 2D or 3D. If any/all of the parameters are None, default
vales are assigned.
Parameters
----------
image_or_array : array_like, itk.Image, or vtk.vtkImageData
The 2D or 3D image to visualize.
point1 : list of float
List elements represent the 2D/3D coordinate of the point1.
point2 : list of float
List elements represent the 2D/3D coordinate of the point2.
order : int, optional
Spline order for line profile interpolation. The order has to be in the
range 0-5.
"""
if not have_scipy:
raise RuntimeError(
'The scipy package in necessary for the line_profiler widget.')
if not have_itk:
raise RuntimeError(
'The itk package in necessary for the line_profiler widget.')
if image_or_array is None:
image_or_array = self.image
if point1 is None:
point1 = self.point1
if point2 is None:
point2 = self.point2
if order is None:
order = self.order
image = to_itk_image(image_or_array)
image_array = itk.array_view_from_image(image)
dimension = image.GetImageDimension()
distance = np.sqrt(
sum([(point1[ii] - point2[ii])**2 for ii in range(dimension)]))
index1 = tuple(
image.TransformPhysicalPointToIndex(tuple(point1[:dimension])))
index2 = tuple(
image.TransformPhysicalPointToIndex(tuple(point2[:dimension])))
num_points = int(
np.round(
np.sqrt(
sum([(index1[ii] - index2[ii])**2
for ii in range(dimension)])) * 2.1))
coords = [
np.linspace(index1[ii], index2[ii], num_points)
for ii in range(dimension)
]
mapped = scipy.ndimage.map_coordinates(image_array,
np.vstack(coords[::-1]),
order=order,
mode='nearest')
return np.linspace(0.0, distance, num_points), mapped
class Viewer(ViewerParent):
"""Viewer widget class."""
_view_name = Unicode('ViewerView').tag(sync=True)
_model_name = Unicode('ViewerModel').tag(sync=True)
_view_module = Unicode('itk-jupyter-widgets').tag(sync=True)
_model_module = Unicode('itk-jupyter-widgets').tag(sync=True)
_view_module_version = Unicode('^0.15.2').tag(sync=True)
_model_module_version = Unicode('^0.15.2').tag(sync=True)
image = ITKImage(default_value=None,
allow_none=True,
help="Image to visualize.").tag(sync=False,
**itkimage_serialization)
rendered_image = ITKImage(default_value=None,
allow_none=True).tag(sync=True,
**itkimage_serialization)
_rendering_image = CBool(
default_value=False,
help="We are currently volume rendering the image.").tag(sync=True)
ui_collapsed = CBool(
default_value=False,
help="Collapse the built in user interface.").tag(sync=True)
annotations = CBool(default_value=True,
help="Show annotations.").tag(sync=True)
mode = CaselessStrEnum(
('x', 'y', 'z', 'v'),
default_value='v',
help="View mode: x: x plane, y: y plane, z: z plane, v: volume rendering"
).tag(sync=True)
interpolation = CBool(
default_value=True,
help="Use linear interpolation in slicing planes.").tag(sync=True)
cmap = Unicode('Viridis (matplotlib)').tag(sync=True)
shadow = CBool(
default_value=True,
help="Use shadowing in the volume rendering.").tag(sync=True)
slicing_planes = CBool(
default_value=False,
help="Display the slicing planes in volume rendering view mode.").tag(
sync=True)
gradient_opacity = CFloat(
default_value=0.2,
help="Volume rendering gradient opacity, from (0.0, 1.0]").tag(
sync=True)
roi = NDArray(dtype=np.float64, default_value=np.zeros((2, 3), dtype=np.float64),
help="Region of interest: ((lower_x, lower_y, lower_z), (upper_x, upper_y, upper_z))")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(2, 3))
_largest_roi = NDArray(dtype=np.float64, default_value=np.zeros((2, 3), dtype=np.float64),
help="Largest possible region of interest: ((lower_x, lower_y, lower_z), (upper_x, upper_y, upper_z))")\
.tag(sync=True, **array_serialization)\
.valid(shape_constraints(2, 3))
select_roi = CBool(
default_value=False,
help="Enable an interactive region of interest widget for the image."
).tag(sync=True)
size_limit_2d = NDArray(
dtype=np.int64,
default_value=np.array([1024, 1024], dtype=np.int64),
help="Size limit for 2D image visualization.").tag(sync=False)
size_limit_3d = NDArray(
dtype=np.int64,
default_value=np.array([192, 192, 192], dtype=np.int64),
help="Size limit for 3D image visualization.").tag(sync=False)
_downsampling = CBool(
default_value=False,
help="We are downsampling the image to meet the size limits.").tag(
sync=True)
_reset_crop_requested = CBool(
default_value=False,
help="The user requested a reset of the roi.").tag(sync=True)
def __init__(self, **kwargs):
super(Viewer, self).__init__(**kwargs)
dimension = self.image.GetImageDimension()
largest_region = self.image.GetLargestPossibleRegion()
size = largest_region.GetSize()
# Cache this so we do not need to recompute on it when resetting the roi
self._largest_roi_rendered_image = None
self._largest_roi = np.zeros((2, 3), dtype=np.float64)
if not np.any(self.roi):
largest_index = largest_region.GetIndex()
self.roi[0][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index))
largest_index_upper = largest_index + size
self.roi[1][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index_upper))
self._largest_roi = self.roi.copy()
if dimension == 2:
for dim in range(dimension):
if size[dim] > self.size_limit_2d[dim]:
self._downsampling = True
else:
for dim in range(dimension):
if size[dim] > self.size_limit_3d[dim]:
self._downsampling = True
if self._downsampling:
self.extractor = itk.ExtractImageFilter.New(self.image)
self.extractor.InPlaceOn()
self.shrinker = itk.BinShrinkImageFilter.New(self.extractor)
self._update_rendered_image()
if self._downsampling:
self.observe(self._on_roi_changed, ['roi'])
self.observe(self._on_reset_crop_requested, ['_reset_crop_requested'])
self.observe(self.update_rendered_image, ['image'])
@debounced(delay_seconds=1.5, method=True)
def _on_roi_changed(self, change=None):
if self._downsampling:
self._update_rendered_image()
def _on_reset_crop_requested(self, change=None):
if change.new == True and self._downsampling:
dimension = self.image.GetImageDimension()
largest_region = self.image.GetLargestPossibleRegion()
size = largest_region.GetSize()
largest_index = largest_region.GetIndex()
new_roi = self.roi.copy()
new_roi[0][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index))
largest_index_upper = largest_index + size
new_roi[1][:dimension] = np.array(
self.image.TransformIndexToPhysicalPoint(largest_index_upper))
self._largest_roi = new_roi.copy()
self.roi = new_roi
if change.new == True:
self._reset_crop_requested = False
@debounced(delay_seconds=0.2, method=True)
def update_rendered_image(self, change=None):
self._largest_roi_rendered_image = None
self._largest_roi = np.zeros((2, 3), dtype=np.float64)
self._update_rendered_image()
@staticmethod
def _find_shrink_factors(limit, dimension, size):
shrink_factors = [
1,
] * dimension
for dim in range(dimension):
while (int(np.floor(float(size[dim]) / shrink_factors[dim])) >
limit[dim]):
shrink_factors[dim] += 1
return shrink_factors
def _update_rendered_image(self):
if self.image is None:
return
if self._rendering_image:
@yield_for_change(self, '_rendering_image')
def f():
x = yield
assert (x == False)
f()
self._rendering_image = True
if self._downsampling:
dimension = self.image.GetImageDimension()
index = self.image.TransformPhysicalPointToIndex(
self.roi[0][:dimension])
upper_index = self.image.TransformPhysicalPointToIndex(
self.roi[1][:dimension])
size = upper_index - index
if dimension == 2:
shrink_factors = self._find_shrink_factors(
self.size_limit_2d, dimension, size)
else:
shrink_factors = self._find_shrink_factors(
self.size_limit_3d, dimension, size)
self.shrinker.SetShrinkFactors(shrink_factors)
region = itk.ImageRegion[dimension]()
region.SetIndex(index)
region.SetSize(tuple(size))
# Account for rounding
# truncation issues
region.PadByRadius(1)
region.Crop(self.image.GetLargestPossibleRegion())
self.extractor.SetExtractionRegion(region)
size = region.GetSize()
is_largest = False
if np.any(self._largest_roi) and np.all(
self._largest_roi == self.roi):
is_largest = True
if self._largest_roi_rendered_image is not None:
self.rendered_image = self._largest_roi_rendered_image
return
self.shrinker.UpdateLargestPossibleRegion()
if is_largest:
self._largest_roi_rendered_image = self.shrinker.GetOutput()
self._largest_roi_rendered_image.DisconnectPipeline()
self._largest_roi_rendered_image.SetOrigin(
self.roi[0][:dimension])
self.rendered_image = self._largest_roi_rendered_image
return
shrunk = self.shrinker.GetOutput()
shrunk.DisconnectPipeline()
shrunk.SetOrigin(self.roi[0][:dimension])
self.rendered_image = shrunk
else:
self.rendered_image = self.image
@validate('gradient_opacity')
def _validate_gradient_opacity(self, proposal):
"""Enforce 0 < value <= 1.0."""
value = proposal['value']
if value <= 0.0:
return 0.01
if value > 1.0:
return 1.0
return value
@validate('cmap')
def _validate_cmap(self, proposal):
value = proposal['value']
if not value in COLORMAPS:
raise TraitError('Invalid colormap')
return value
def roi_region(self):
"""Return the itk.ImageRegion corresponding to the roi."""
dimension = self.image.GetImageDimension()
index = self.image.TransformPhysicalPointToIndex(
tuple(self.roi[0][:dimension]))
upper_index = self.image.TransformPhysicalPointToIndex(
tuple(self.roi[1][:dimension]))
size = upper_index - index
for dim in range(dimension):
size[dim] += 1
region = itk.ImageRegion[dimension]()
region.SetIndex(index)
region.SetSize(tuple(size))
region.Crop(self.image.GetLargestPossibleRegion())
return region
def roi_slice(self):
"""Return the numpy array slice corresponding to the roi."""
dimension = self.image.GetImageDimension()
region = self.roi_region()
index = region.GetIndex()
upper_index = np.array(index) + np.array(region.GetSize())
slices = []
for dim in range(dimension):
slices.insert(0, slice(index[dim], upper_index[dim] + 1))
return tuple(slices)