Reconstruct a TomoStack series using one of the available methods.

Parameters:¶

method: 'FBP' | 'SIRT' | 'SART' | 'DART' = 'FBP'¶

Reconstruction algorithm to use. Must be one of "FBP" (default), "SIRT", "SART", or "DART"

iterations: int = 5¶

Number of iterations for the SIRT reconstruction (used with SIRT, SART, and DART methods) (default: 5)

constrain: bool = False¶

If True, output reconstruction is constrained above value given by thresh

thresh: float = 0¶

Value above which to constrain the reconstructed data

cuda: bool | None = None¶

Whether or not to use CUDA-accelerated reconstruction algorithms. If None (the default), the decision to use CUDA will be left to astra.astra.use_cuda().

thickness: int | None = None¶

Size of the output volume (in pixels) in the projection direction. If None, the y-size of the stack is used.

show_progressbar: bool = True¶

If True, show a progress bar for the reconstruction. Default is True.

p: float = 0.99¶

Probability for setting free pixels in DART reconstruction (only used if the reconstruction method is DART, default: 0.99)

ncores: int | None = None¶

Number of cores to use for multithreaded reconstructions.

sino_filter: 'ram-lak' | 'shepp-logan' | 'cosine' | 'hamming' | 'hann' | 'none' | 'tukey' | 'lanczos' | 'triangular' | 'gaussian' | 'barlett-hann' | 'blackman' | 'nuttall' | 'blackman-harris' | 'blackman-nuttall' | 'flat-top' | 'kaiser' | 'parzen' | 'projection' | 'sinogram' | 'rprojection' | 'rsinogram' = 'shepp-logan'¶

Filter for filtered backprojection. Default is "shepp-logan". Available options are detailed in the Astra Toolbox documentation under the cfg.FilterType option of FBP_CUDA.

dart_iterations: int | None = 5¶

Number of iterations to employ for DART reconstruction

gray_levels: list | ndarray | None = None¶

List of gray levels to use for DART reconstruction

Returns:¶

rec – RecStack containing the reconstructed volume

Return type:¶

RecStack

Examples¶

Filtered backprojection (FBP) reconstruction:

>>> import etspy.datasets as ds
>>> stack = ds.get_needle_data(aligned=True)
>>> slices = stack.isig[:, 120:121].deepcopy()
>>> rec = slices.reconstruct('FBP', cuda=False, show_progressbar=False)

Simultaneous iterative reconstruction technique (SIRT) reconstruction:

>>> import etspy.datasets as ds
>>> stack = ds.get_needle_data(aligned=True)
>>> slices = stack.isig[:, 120:121].deepcopy()
>>> rec = slices.reconstruct('SIRT',iterations=5,
...                          cuda=False, show_progressbar=False)

SIRT reconstruction with positivity constraint:

>>> import etspy.datasets as ds
>>> stack = ds.get_needle_data(aligned=True)
>>> slices = stack.isig[:, 120:121].deepcopy()
>>> iterations = 5
>>> constrain = True
>>> thresh = 0
>>> rec = slices.reconstruct('SIRT', iterations, constrain, thresh,
...                          cuda=False, show_progressbar=False)

Discreate algebraice reconstruction technique (DART) reconstruction:

>>> import etspy.datasets as ds
>>> stack = ds.get_needle_data(aligned=True)
>>> slices = stack.isig[:, 120:121].deepcopy()
>>> gray_levels = [0., slices.data.max()/2, slices.data.max()]
>>> rec = slices.reconstruct('DART', iterations=5, cuda=False,
...                          gray_levels=gray_levels, p=0.99,
...                          dart_iterations=5, show_progressbar=False)