def get_ImageData(num_model, geometry):
'''Returns an ImageData relative to geometry with the model num_model from tomophantom
:param num_model: model number
:type num_model: int
:param geometry: geometrical info that describes the phantom
:type geometry: ImageGeometry
Example usage:
.. code-block:: python
ndim = 2
N=128
angles = np.linspace(0, 360, 50, True, dtype=np.float32)
offset = 0.4
channels = 3
if ndim == 2:
ag = AcquisitionGeometry.create_Cone2D((offset,-100), (offset,100))
ag.set_panel(N)
else:
ag = AcquisitionGeometry.create_Cone3D((offset,-100, 0), (offset,100,0))
ag.set_panel((N,N-2))
ag.set_channels(channels)
ag.set_angles(angles, angle_unit=AcquisitionGeometry.DEGREE)
ig = ag.get_ImageGeometry()
num_model = 1
phantom = TomoPhantom.get_ImageData(num_model=num_model, geometry=ig)
'''
ig = geometry.copy()
ig.set_labels(DataOrder.TOMOPHANTOM_IG_LABELS)
num_dims = len(ig.dimension_labels)
if ImageGeometry.CHANNEL in ig.dimension_labels:
if not is_model_temporal(num_model):
raise ValueError('Selected model {} is not a temporal model, please change your selection'.format(num_model))
if num_dims == 4:
# 3D+time for tomophantom
# output dimensions channel and then spatial,
# e.g. [ 'channel', 'vertical', 'horizontal_y', 'horizontal_x' ]
num_model = num_model
shape = tuple(ig.shape[1:])
phantom_arr = TomoP3D.ModelTemporal(num_model, shape, path_library3D)
elif num_dims == 3:
# 2D+time for tomophantom
# output dimensions channel and then spatial,
# e.g. [ 'channel', 'horizontal_y', 'horizontal_x' ]
N = ig.shape[1]
num_model = num_model
phantom_arr = TomoP2D.ModelTemporal(num_model, ig.shape[1], path_library2D)
else:
raise ValueError('Wrong ImageGeometry')
if ig.channels != phantom_arr.shape[0]:
raise ValueError('The required model {} has {} channels. The ImageGeometry you passed has {}. Please update your ImageGeometry.'\
.format(num_model, ig.channels, phantom_arr.shape[0]))
else:
if num_dims == 3:
# 3D
num_model = num_model
phantom_arr = TomoP3D.Model(num_model, ig.shape, path_library3D)
elif num_dims == 2:
# 2D
if ig.shape[0] != ig.shape[1]:
raise ValueError('Can only handle square ImageData, got shape'.format(ig.shape))
N = ig.shape[0]
num_model = num_model
phantom_arr = TomoP2D.Model(num_model, N, path_library2D)
else:
raise ValueError('Wrong ImageGeometry')
im_data = ImageData(phantom_arr, geometry=ig, suppress_warning=True)
im_data.reorder(list(geometry.dimension_labels))
return im_data
print("MAE" , mae(res1, res2))
np.testing.assert_array_almost_equal(res1.as_array(), res2.as_array(), decimal=3)
###################################################################
###################################################################
###################################################################
###################################################################
print("Compare CIL_FGP_TV vs CCPiReg_FGP_TV no tolerance (3D)")
print ("Building 3D phantom using TomoPhantom software")
model = 13 # select a model number from the library
N_size = 64 # Define phantom dimensions using a scalar value (cubic phantom)
path = os.path.dirname(tomophantom.__file__)
path_library3D = os.path.join(path, "Phantom3DLibrary.dat")
#This will generate a N_size x N_size x N_size phantom (3D)
phantom_tm = TomoP3D.Model(model, N_size, path_library3D)
ig = ImageGeometry(N_size, N_size, N_size)
data = ig.allocate()
data.fill(phantom_tm)
n1 = TestData.random_noise(data.as_array(), mode = 'gaussian', seed = 10)
noisy_data = ig.allocate()
noisy_data.fill(n1)
# Show Ground Truth and Noisy Data
plt.figure(figsize=(10,5))
plt.subplot(1,2,1)
plt.imshow(data.as_array()[32])
plt.title('Ground Truth')
def test_compare_regularisation_toolkit_tomophantom(self):
print("Compare CIL_FGP_TV vs CCPiReg_FGP_TV no tolerance (3D)")
print("Building 3D phantom using TomoPhantom software")
model = 13 # select a model number from the library
N_size = 64 # Define phantom dimensions using a scalar value (cubic phantom)
path = os.path.dirname(tomophantom.__file__)
path_library3D = os.path.join(path, "Phantom3DLibrary.dat")
#This will generate a N_size x N_size x N_size phantom (3D)
phantom_tm = TomoP3D.Model(model, N_size, path_library3D)
ig = ImageGeometry(N_size, N_size, N_size)
data = ig.allocate()
data.fill(phantom_tm)
noisy_data = noise.gaussian(data, seed=10)
alpha = 0.1
iters = 1000
print("Use tau as an array of ones")
# CIL_TotalVariation no tolerance
g_CIL = alpha * TotalVariation(iters, tolerance=None, info=True)
res1 = g_CIL.proximal(noisy_data, ig.allocate(1.))
t0 = timer()
res1 = g_CIL.proximal(noisy_data, ig.allocate(1.))
t1 = timer()
print(t1 - t0)
# CCPi Regularisation toolkit high tolerance
r_alpha = alpha
r_iterations = iters
r_tolerance = 1e-9
r_iso = 0
r_nonneg = 0
r_printing = 0
g_CCPI_reg_toolkit = CCPiReg_FGP_TV(r_alpha, r_iterations, r_tolerance,
r_iso, r_nonneg, r_printing, 'cpu')
t2 = timer()
res2 = g_CCPI_reg_toolkit.proximal(noisy_data, 1.)
t3 = timer()
print(t3 - t2)
np.testing.assert_array_almost_equal(res1.as_array(),
res2.as_array(),
decimal=3)
# CIL_FGP_TV no tolerance
#g_CIL = FGP_TV(ig, alpha, iters, tolerance=None, info=True)
g_CIL.tolerance = None
t0 = timer()
res1 = g_CIL.proximal(noisy_data, 1.)
t1 = timer()
print(t1 - t0)
###################################################################
###################################################################
###################################################################
###################################################################
data = dataexample.PEPPERS.get(size=(256, 256))
ig = data.geometry
ag = ig
noisy_data = noise.gaussian(data, seed=10)
alpha = 0.1
iters = 1000
# CIL_FGP_TV no tolerance
g_CIL = alpha * TotalVariation(iters, tolerance=None)
t0 = timer()
res1 = g_CIL.proximal(noisy_data, 1.)
t1 = timer()
print(t1 - t0)
# CCPi Regularisation toolkit high tolerance
r_alpha = alpha
r_iterations = iters
r_tolerance = 1e-8
r_iso = 0
r_nonneg = 0
r_printing = 0
g_CCPI_reg_toolkit = CCPiReg_FGP_TV(r_alpha, r_iterations, r_tolerance,
r_iso, r_nonneg, r_printing, 'cpu')
t2 = timer()
res2 = g_CCPI_reg_toolkit.proximal(noisy_data, 1.)
t3 = timer()
print(t3 - t2)
