Z-stitching

Since nabu 2024.1.0 nabu provides an API to stitch volume and / or projections

nabu-stitching-config will help you create a configuration for stitching (like nabu-config for tomographic reconstruction

nabu-stitching will execute the configuration

introduction

z-stitching can be done over:

• projections (in this case we will speak of pre-processing stitching)
• reconstructed volumes (in this case we will speak about post-processing stitching)

At nabu level most of the code is shared for those two use-case. Diversity comes mostly from getting and interpretting metadata of the scans or of the volumes.

overlaping

To process successfully a stitching frames must have an overlaping area

pre processing z-stitching (projections)

You can stitch together a z-serie in pre processing by using the PreProcessZStitcher. It requires as input a set of Nxtomo.

real life example of z-stitching projections

reference used for projection stitching

post processing z-stitching (reconstructed volumes)

You can stitch together a z-serie in pre processing by using the PostProcessZStitcher. It requires as input a set of (reconstructed) volumes.

real life example of z-stitching volumes

reference used for volume stitching

stitching using the GUI

A GUI exists to stitch volumes with a graphical user interface. It is embed with tomwer. See TODO: add link

stitching from the command line interface

You can apply stitching from CLI by:

• creating a configuration file
• editing the configuration
• calling nabu-stitching with the configuration file

creating a configuration file

To create a configuration file you can use the nabu-stitching-config command

Please mind using the --datasets options to speed up the configuration file creation

# ! nabu-stitching-config --help

Exercise A:

A.1 create a configuration file with NXtomos contained in /data/projects/tomo-sample-data/training/z-stitching/STYLO/PROCESSED_DATA/nxtomos

A.2: edit the configuration to stitch only one slice of the volume

[inputs]
slices = 200

A.3 provide the output path

txt
[z-preproc]
# output nxtomo file path
location = stitching/stitching.nx
# output nxtomo data path
data_path = my_entry

Executing a stitching

The processing of stitching configuration is done by calling nabu-stitching over a stuitching configuration file like:

nabu-stitching [my_stitching_file]

Exercise B:

Execute the stitching configuration created in Exercise A

Warning: first execution execution time: During the first execution of the stitching of NXtomo it will create all reduced_darks and reduced_flats. So this calculation can take some time.

Then display the result using silx view

Stitching kernels

For the overlap area you can select between several kernels / strategies

You can define this parameter in

txt
[stitching]
# Policy to apply to compute the overlap area. Must be in ('mean', 'cosinus weights', 'linear weights', 'closest', 'image minimum divergence', 'higher signal').
stitching_strategy = cosinus weights

Exercise C: modify the stitching_strategy of the configuration file and run it again

Distributing

When you have a large dataset in order to speed up the stitching you can distribute the stitching over several slurm job.

This distribution is defined in the [slurm] section. If the slurm section is not present or if the partition field is empty then the stitching will be done locally.

Else it will be split into n_jobs. Each job will create is own part and once all job are finished a master file will be created linking all the sub part.

If you are quicked out before the master file could be created you can call nabu-stitching with the --only-create-master-file option to only process the master file creation.

Exercise D: distribute some stitching

• reconstruct 20 slices

txt
[inputs]
slices = 100:120

• with 2 jobs

txt
[slurm]
# section didicated to slurm. If you want to run locally avoid setting 'partition or remove this section'
# slurm partition to be used. If empty will run locally
partition = low
# number of job to launch (split computation on N parallel jobs). Once all are finished we will concatenate the result.
n_jobs = 2

• use the module tomotools/training_2024

txt
[slurm]
# module to load
modules = tomotools/training_2024