nabu.stitching.stitcher.single_axis module

class nabu.stitching.stitcher.single_axis.SingleAxisStitcher(configuration, *args, **kwargs)

Bases: _StitcherBase

Any single-axis base class

property axis: int

property dumper

property stitching_axis_in_frame_space

//tomo.gitlab-pages.esrf.fr/bliss-tomo/master/modelization.html)

Type:

stitching is operated in 2D (frame) space. So the axis in frame space is different than the one in 3D ebs-tomo space (https

stitch(store_composition: bool = True) → BaseIdentifier

Apply expected stitch from configuration and return the DataUrl of the object created

Parameters:

store_composition (bool) – if True then store the composition used for stitching in frame_composition. So it can be reused by third part (like tomwer) to display composition made

property serie_label: str

return serie name for logs

get_n_slices_to_stitch()

Return the number of slice to be stitched

get_final_axis_positions_in_px() → dict

compute the final position (in pixel) from the initial position of the first object and the final relative shift computed (1) (1): the final relative shift is obtained from the initial shift (from motor position of provided by the user) + the refinement shift from cross correlation algorithm :return: dict with tomo object identifier (str) as key and a tuple of position in pixel (axis_0_pos, axis_1_pos, axis_2_pos)

settle_flips()

User can provide some information on existing flips at frame level. The goal of this step is to get one flip_lr and on flip_ud value per scan or volume

property series: Series

property configuration: SingleAxisStitchingConfiguration

property progress

rescale_frames(frames: tuple)

rescale_frames if requested by the configuration

normalize_frame_by_sample(frames: tuple)

normalize frame from a sample picked on the left or the right

static stitch_frames(frames: tuple | ndarray, axis, x_relative_shifts: tuple, y_relative_shifts: tuple, output_dtype: ndarray, stitching_axis: int, overlap_kernels: tuple, dumper: DumperBase = None, check_inputs=True, shift_mode='nearest', i_frame=None, return_composition_cls=False, alignment='center', pad_mode='constant', new_width: int | None = None) → ndarray

shift frames according to provided shifts (as y, x tuples) then stitch all the shifted frames together and save them to output_dataset.

Parameters:

• frames (tuple) – element must be a DataUrl or a 2D numpy array

• stitching_regions_hdf5_dataset –