Skip to content

Latest commit

 

History

History
214 lines (155 loc) · 11 KB

README.md

File metadata and controls

214 lines (155 loc) · 11 KB
      %&&&&&+   +&&&&&&&*+                                                                 
      #@@@@@+  *#@@@@@#                                                                 *+ 
     *@@#*@@% *@@@%%@@+   %#&&@#%   &##&&#&%  +#@#&&##%+ +&&    &@@* &###&%%%*  +%&+   &@% 
    +@@@% &@@%@@#+ #@%  %@@*   @@& +@@%  +@@%  @@&  +@@% &@%   +@@%  &@#+      +@@@&  *@#  
    %@@#  +@@@@#  %@@  %@@*    @@& &@@&&&#&%  %@@%*%@#% &@@##&&@@@  +@@#&&%*+  #@#&@& #@*  
   +@@@*   &@@#   #@%  *@@%   &@#++@@%**#@%   @@&**+    &@*   %@@%  &@@+      &@@  *@#@&   
    %@&    *@&   &#&    +&##&#&%  &#&   *@@% %##+      *#%    ##&  %####&%%%*%@@&   *##+   
            +    +                       +%@&%***                          +&@@&+          
                     ++**%%%%%%%***++      +***+                         +%@@#%            
             +*%%&#@@@#&&%%%%%%%%&&#@@###@@@@@@#*                      *&@@#*              
         +*&#@@@#&%*+                *@@@@#&&@@@@                  +%#@@#%+                
      +&#@@@&*+                      #@@@*+  #@@@&%*++    +++**%&#@@#&*                    
    +&@@@&*                          &@@@@@@@@@@%%&####@@@@@##&&%*+                        
   %@@@%+                             +%&&&##&%+                                           
  %@@#+                                                                                    
  %&+                                                                                   

morphen: Overview, features and limitations

Readme file under development!

What is morphen?

morphen is a collection of python-based astronomical functionalities for image analysis and processing. The three main functionalities of morphen are:

  1. Image Analysis (image_analysis/README.md)
  2. Multi-Sersic Image Fitting Decomposition (image_decomposition/README.md)
  3. Radio Interferometric Self-calibration (selfcal/README.md)

You will be able to measure basic image morphology and photometry. morphen also comes with a state-of-the-art python-based image fitting implementation based on the Sersic function. Particularly to radio astronomy, these tools involve pure python, but also are integrated with CASA (https://casa.nrao.edu/) in order to work with common casatasks as well WSClean -- a code for fast interferometric imaging (https://wsclean.readthedocs.io/en/latest/).

Getting Started

Some specifics of what you can do with morphen includes:

  • Perform morphological analysis of astronomical images (general.)
  • Basic source extraction and photometry (general).
  • Perform a multi-component Sersic image decomposition to astronomical images of galaxies (general).
  • Perform self-calibration and imaging with WSClean and CASA (radio astronomy).
  • Use information from distinct interferometric arrays to perform a joint separation of distinct physical mechanisms of the radio emission (radio astronomy).
  • Experimental: some functionalities are applicable to general astronomical data, but more testing is required beyond radio astronomy.

While in development, these modules will be kept in the same place. Stable releases will be provided for the full module. However, we plan to release these separated functionalities as standalone repositories in the near future.

Currently, there is no option to install morphen (via pip or conda). However, its usage is simple. The code can be used as a module, interactively via Jupyter notebooks, or via the command line interface (see "Important notes" below). For now, we recommend
using it via Jupyter notebooks (see below for examples).

The modular file morphen.py is the on-development module that allows you to do such tasks, like a normal package installed via pip. For that, need to download the entire repository. The libs directory, specifically the libs/libs.py file, contains the core functionalities in which morphen.py is based. Examples can be found in the following directories:

  • notebooks/: contains some more general examples of how to use the code.
  • image_analysis/: contains examples of how to use the image analysis functionalities.
  • image_decomposition/: contains examples of how to use the Sersic image decomposition functionalities.

Important notes

  1. The functionalities presented in the examples notebooks are stable. We are in extensive development, and we are setting milestones for optimizations, bug fixes, better documentation, and new functionalities for a larger scope of the code.
  2. The command line option is still under development and not all argument options are available. However, using it via jupyter is somehow stable (check the notebooks for examples).
  3. This readme file is under development. I am also currently adding more basic usages to Jupyter notebooks guides.
  4. Installation instructions for all the dependencies are provided in the install_instructions.md file.

Features

Image Analysis

In the directory image_analysis/, the notebook morphen.ipynb contain sets examples of how to perform basic image analysis, such as image statistics, photometry, shape analysis, etc. Check also image_analysis/README.md file for more details.

Image Fitting Decomposition

We introduce a Python-based image fitting implementation using the Sersic function. This implementation is designed to be robust, fast with GPU acceleration using JaX (https://jax.readthedocs.io/en/latest/index.html) and easy to use (semi-automated). The physical motivation behind this implementation is to provide an interface to easily perform a multi-component decomposition constrained around prior knowledge from the data itself, without the need of creating complicated configuration files to set model parameters. This helps mitigate issues when trying to fit multiple-component models to the data. Prior photometry is measured from the data using the PetroFit code (https://petrofit.readthedocs.io/en/latest/index.html) and the photutils package (https://photutils.readthedocs.io/en/stable/) and used as initial constraints for the minimisation. Examples of how to use it can be found in the Notebook image_decomposition/morphen_sersic.ipynb

The decomposition was first designed for radio interferometric images, but can be used with any other type of images, such as optical images. However, application to optical data is still a work in progress as we require better PSF modeling, especially for HST and JWST observations. For now, you already can check some basic examples in the notebook image_decomposition/morphen_sersic_optical.ipynb

Radio Interferometric Related Tasks

Interferometric Imaging With wsclean

Directory imaging/ contains a python script called imaging/imaging_with_wsclean_v3.py which is just a support code for easy use to call wsclean on the command line. See the intructions file of how to use it: imaging/wsclean_imaging.md

Imaging with wsclean and self-calibration

File selfcal/imaging_with_wsclean.py is a wrapper to call wsclean on the command line, with pre-defined parameters already set. You can use it to perform imaging with wsclean in a simple way and change parameters as required. Note that not all WSClean arguments are available in this wrapper. Arguments that are not implemented can be simply passed with the argument --opt_args in imaging_with_wsclean.py. This script is standalone and can be downloaded and used separately from the morphen package.

In previous versions of this module (not available in this repo), all self-calibration routines were done with CASA. However, some changes were made and in this repo we provide for the first time an automated way to perform self-calibration, which uses WSClean as imager and CASA to compute the complex gain corrections (phases and amplitudes).

To check how to use it, see the selfcal/README.md file and examples in selfcal/selfcalibration.ipynb. This self-calibration pipeline was tested in multiple datasets with the VLA from 1.4 GHz to 33 GHz and with e-MERLIN at 5 GHz, for a wide range of sources total flux densities.

The file selfcal/auto_selfcal_wsclean.py is a script to perform self-calibration with wsclean and CASA. Is fully automated, but is still in development. Check the selfcal/README.md file for more details.

Selfcalibration and Imaging with CASA.

(DOC NOT READY)

Interferometric Decomposition

Interferometric decomposition is a technique introduced by Lucatelli et al. (2024) to disentangle the radio emission using combined images from distinct interferometric arrays.

More details will be provided soon.

CASA Utilities

(IN DEV)

Origin of morphen

The idea of morphen predates back to 2018 alongside the development of morfometryka (https://iopscience.iop.org/article/10.1088/0004-637X/814/1/55/pdf) and $$\kappa$$urvature (https://academic.oup.com/mnras/article/489/1/1161/5543965). The aim was to expand the functionalities of morfometryka (such as automated bulge-disk decomposition) and some optimisations. Development was on pause, but soon after I started working with radio astronomy, it was clear that we needed a set of automated tools for radio interferometric data processing and analysis, from basic plotting to more complicated tasks, such as self-calibration and a robust image decomposition.

Alongside, it was also clear that the reproducibility in radio astronomy is a challenge, and we were in need of a package towards reproducibility of scientific results. Hence, the ideas of morphen were brought back to be incorporated within radio astronomy.

How to contribute

This is an open-source project. We are welcoming all kinds of contributions, suggestions and bug reports. Feel free to open an issue or contact us.