Laminocupy-cli¶

Laminocupy-cli is a command-line interface for GPU reconstruction of laminographic data. All preprocessing operations are implemented on GPU with using cupy library.

Features¶

List here
the module features

Contribute¶

Documentation: https://laminocupy.readthedocs.io/en/latest/#
Issue Tracker: https://github.com/nikitinvv/laminocupy-cli/issues
Source Code: https://github.com/nikitinvv/laminocupy-cli/

Content¶

Install¶

Create environment with necessary dependencies

(base)$ conda create -n laminocupy -c conda-forge python=3.9 dxchange cupy scikit-build swig pywavelets numexpr astropy olefile opencv
(base)$ conda activate laminocupy
(laminocupy)$ pip install torch==1.9.1+cu111 torchvision==0.10.1+cu111 torchaudio==0.9.1 -f https://download.pytorch.org/whl/torch_stable.html

Install the pytorch pywavelets package for ring removal

(laminocupy)$ git clone https://github.com/fbcotter/pytorch_wavelets
(laminocupy)$ cd pytorch_wavelets
(laminocupy)$ pip install .
(laminocupy)$ cd -

Set path to the nvcc profiler (e.g. /local/cuda-11.4/bin/nvcc ) and install laminocupy

(laminocupy)$ export CUDACXX=/local/cuda-11.4/bin/nvcc
(laminocupy)$ git clone https://github.com/nikitinvv/laminocupy-cli
(laminocupy)$ cd laminocupy-cli
(laminocupy)$ python setup.py install

Update¶

laminocupy-cli is constantly updated to include new features. To update your locally installed version:

(laminocupy)$ cd laminocupy-cli
(laminocupy)$ git pull
(laminocupy)$ python setup.py install

Usage¶

Example¶

(laminocupy)$ laminocupy reconstep --file-name /data/2021-12/Duchkov/exp4_ho_130_vertical_0_2018.h5 --remove-stripe-method fw --nproj-per-chunk 32 --nsino-per-chunk 32 --reconstruction-type full --rotation-axis 1198 --lamino-angle 30

More options¶

(laminocupy)$ laminocupy -h
(laminocupy)$ laminocupy reconstep -h

Performance¶

API reference¶

laminocupy_cli Modules:

`laminocupy_cli.rec_steps`¶

`laminocupy_cli.remove_stripe`¶

`laminocupy_cli.retrieve_phase`¶

Credits¶

Citations¶

We kindly request that you cite the following article [A1] if you use laminocupy-cli

A1: Viktor Nikitin, Aniket Tekawade, Anton Duchkov, Pavel Shevchenko, and Francesco De Carlo. Real-time streaming tomographic reconstruction with on-demand data capturing and 3d zooming to regions of interest. Journal of Synchrotron Radiation, 2022.

References¶

B1: Viktor Nikitin, Aniket Tekawade, Anton Duchkov, Pavel Shevchenko, and Francesco De Carlo. Real-time streaming tomographic reconstruction with on-demand data capturing and 3d zooming to regions of interest. Journal of Synchrotron Radiation, 2022.