Skip to main content

Grid-Free Deconvolution Directly From Visibilities

Project description

Spotless - Radio Interferometry Imaging Algorithm

This is a point-source deconvolution algorithm (part of the CLEAN family) that works without gridding. [http://www.iram.fr/IRAMFR/GILDAS/doc/html/map-html/node37.html]

This is essentially a grid-free version of the Cotton-Schwab algorithm with a different convergence an optimization steps. [Relaxing the isoplanatism assumption in self-calibration; applications to low-frequency radio interferometry]

It does not require W-projection and handles non-coplanar antennas without difficulty. It also works on all-sky images just fine.

Spotless works in a similar way to Hogbom's CLEAN algorithm, however

![Dirty, Spotless and Residual][tart_image]

For more information see the TART Github repository

Install Instructions

tart_tools is available from standard python package repositories. Try:

pip install spotless

Running it on live data

spotless --api https://tart.elec.ac.nz/signal --display --show-sources
gridless --api https://tart.elec.ac.nz/signal --display --show-sources

TODO

  • Add Gaussian Source Model
  • Make explicit the antenna model (gain as a function of angular coordinates). We are assuming it is hemispherical here.
  • Prove the relationship between power in the image and visibilty amplitudes. This might only work when the image tends towards a random one. But this is OK since as we remove the sources the residual becomes more and more random.
  • Run an MCMC on the multimodel option to estimate uncertainty in the model. Then use this uncertainty as a stopping criterion (when new model components no longer have certain amplitude or position)

Author

Development work

If you are developing this package, this should be installed using

	make develop

in which case changes to the source-code will be immediately available to projects using it.

Changes

  • 0.4.1 Use the disko sphere. Clean up unused code. Use harmonics from disko. Specify the --fov and --res as in disko
  • 0.4.0 Move to github repository. Add to pypi. Use disko for utility functions. Add a --version CLI argument
  • 0.3.0 Update to python3
  • 0.3.3 Add a gridless binary to plot a GRIDLESS imaging, add a PDF export option
  • 0.3.4 Fix bitrot in multispotless.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

spotless-0.4.1.tar.gz (27.2 kB view details)

Uploaded Source

File details

Details for the file spotless-0.4.1.tar.gz.

File metadata

  • Download URL: spotless-0.4.1.tar.gz
  • Upload date:
  • Size: 27.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.1 CPython/3.10.8

File hashes

Hashes for spotless-0.4.1.tar.gz
Algorithm Hash digest
SHA256 05da3b413a799d8e4f2e5b49df9dda8d5ffe99765cc80408e44ac3f7970f1912
MD5 c3c5972e7018e6c7cf7bbc89cfc22648
BLAKE2b-256 c4619283b73f970ac9af4055d5561309d8faa344e66fe6edd9483ec223849b3e

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page