The main function of this repository is fan_trans(), which can be found in wavan.py
The function fan_trans() performs the Fan wavelet transform as described by Kirby 2005.
Additional options allow the user to perform a Multiscale non-Gaussian segmentation as descirbed by Robitaille et al. 2014 and Robitaille et al. 2019.
Here is a non-hexaustive description for some other functions:
-
The function powspec() calculates the classical Fourier Power Spectrum of an image where the Fourier coefficients are averaged over the azimuthal angles as a function of the wavenumber.
-
The function gauss_beam() allows one to model the transfer function of a telescope by a Gaussian function.
-
Functions subfits(), congrid() (in rebin.py) and imsmooth() allow respectively to cut a fits file, rebin a map and smooth a map.
-
Finally, the function fbm2d() in noisegen.py allows one to create a fractional Brownian motion map, i.e. a fractal image, respecting a specific power law.
powspec(), fbm2d() and apodize() functions are inspired by the mamdlib IDL library.
Dependencies
pip install tqdm
pip install astropy
pywavan installation
Clone this repository:
git clone https://github.com/jfrob27/pywavan.git
and in the pywavan directory
pip install .
or directly
pip install git+https://github.com/jfrob27/pywavan.git
An introduction Notebook tutorial is available here and in the tutorial folder.
Pywavan is also demonstrated on real astrophysical data along some Turbustat functions in the following Notebook. This Notebook is also designed to work directly in Google Colab.