scikit-surgerycalibration

scikit-surgerycalibration provides algorithms designed for the calibration of surgical instruments


Keywords
medical, imaging, aruco-markers, camera-calibration, charuco, opencv, scikit-surgery
License
DSDP
Install
pip install scikit-surgerycalibration==0.2.5

Documentation

scikit-surgerycalibration

Logo

GitHub Actions CI statuss

Test coverage

Documentation Status

The SciKit-Surgery paper

Follow scikit_surgery on twitter

Author(s): Stephen Thompson; Contributor(s): Matt Clarkson, Thomas Dowrick and Miguel Xochicale

scikit-surgerycalibration is part of the SciKit-Surgery software project, developed at the Wellcome EPSRC Centre for Interventional and Surgical Sciences, part of University College London (UCL).

scikit-surgerycalibration is tested on Python 3.7.

scikit-surgerycalibration contains algorithms to perform calibrations useful during surgery, for example pointer calibration, ultrasound calibration, and camera calibration.

Please explore the project structure, and request or implement your desired functionality.

Features

  • Pivot Calibration for pivot calibration.
  • Calibration of mono or stereo tracked video data, calculating camera intrinsics and handeye transformation.

Cloning

You can clone the repository using the following command: :

git clone https://github.com/SciKit-Surgery/scikit-surgerycalibration
git clone git@github.com:SciKit-Surgery/scikit-surgerycalibration.git # Alternatively, use password-protected SSH key.

Developing

We recommend using anaconda or miniconda to create a python 3.7 environment, then using tox to install all dependencies inside a dedicated venv. We then use github actions to run a matrix of builds for Windows, Linux and Mac and various python versions.

All library dependencies are specified via requirements-dev.txt which refers to requirements.txt.

So, assuming either anaconda or miniconda is installed, and your current working directory is the root directory of this project: :

conda create --name scikit-surgery python=3.7
conda activate scikit-surgery
pip install tox
tox

As the tox command runs, it will install all dependencies in a sub-directory .tox/py37 (Linux/Mac) or .tox\py37 (Windows). tox will also run pytest and linting for you.

To run commands inside the same environment as tox, you should: :

source .tox/py37/bin/activate

on Linux/Mac, or if you are Windows user: :

.tox\py37\Scripts\activate

Then you can run pytest, linting, or directly run python scripts, and know that the environment was created correctly by tox.

Generating documentation

The simplest way is again using tox. :

tox -e docs

then open docs/build/html/index.html in your browser.

Running tests

Pytest is used for running unit tests: :

python -m pytest
pytest -v -s tests/algorithms/test_triangulate.py #example for individual tests

Linting

This code conforms to the PEP8 standard. Pylint can be used to analyse the code: :

pylint --rcfile=tests/pylintrc sksurgerycalibration

Installing

You can pip install directly from the repository as follows: :

pip install git+https://github.com/SciKit-Surgery/scikit-surgerycalibration 

Contributing

Please see the contributing guidelines.

Copyright 2020 University College London. scikit-surgerycalibration is released under the BSD-3 license. Please see the license file for details.

Acknowledgements

Supported by Wellcome and EPSRC.