A port of the Dual-Tree Complex Wavelet Transform MATLAB toolbox.
This library provides support for computing 1D, 2D and 3D dual-tree complex wavelet transforms and their inverse in Python. Full documentation is available online.
The easiest way to install dtcwt is via easy_install or pip:
$ pip install dtcwt
If you want to check out the latest in-development version, look at the project’s GitHub page. Once checked out, installation is based on setuptools and follows the usual conventions for a Python project:
$ python setup.py install
(Although the develop command may be more useful if you intend to perform any significant modification to the library.) A test suite is provided so that you may verify the code works on your system:
$ python setup.py nosetests
This will also write test-coverage information to the cover/ directory.
There is more documentation available online and you can build your own copy via the Sphinx documentation system:
$ python setup.py build_sphinx
Compiled documentation may be found in build/docs/html/.
Based on the Dual-Tree Complex Wavelet Transform Pack for MATLAB by Nick Kingsbury, Cambridge University. The original README can be found in ORIGINAL_README.txt. This file outlines the conditions of use of the original MATLAB toolbox.
- Fix regression in dtcwt-based image registration.
- Allow levels used for dtcwt-based image registration to be customised.
- Add queue parameter to low-level OpenCL colifilt and coldfilt functions.
- Significantly increase speed of dtcwt.registration.estimatereg function.
- Fix bug whereby dtcwt.backend_name was not restored when using preserve_backend_stack.
- The OpenCL 2D transform was not always using the correct queue when one was passed explicitly.
- MATLAB-style functions such as dtwavexfm2 have been moved into a separate dtcwt.compat module.
- Backends moved to dtcwt.numpy and dtcwt.opencl modules.
- Removed dtcwt.base.ReconstructedSignal which was a needless wrapper around NumPy arrays.
- Rename TransformDomainSignal to Pyramid.
- Allow runtime configuration of default backend via dtcwt.push_backend function.
- Verified, thanks to @timseries, the NumPy 3D transform implementation against the MATLAB reference implementation.
- Verified the highpass re-sampling routines in dtcwt.sampling against the existing MATLAB implementation.
- Added experimental image registration routines.
- Re-organised documentation.
- Fixed regression from 0.7 where backend_opencl.dtwavexfm2 would return None, None, None.
- Fix a memory leak in OpenCL implementation where transform results were never de-allocated.