search

caracal-pipeline / caracal

Containerized Automated Radio Astronomy Calibration (CARACal) pipeline
GNU General Public License v2.0
29 stars 6 forks source link

Add image-analysis-worker #1406

Open ktrehaeven opened 2 years ago

ktrehaeven commented 2 years ago

Hi

There is a growing concern regarding the manual measurement taking of the flux of radio (mini-) halos. The ever-increasing sensitivity of continuum images and the way in which halo boundaries are manually defined leads to biases when calculating physical measurements. The package Halo-FDCA (Flux Density CAlculator) (https://github.com/JortBox/Halo-FDCA) by Boxelaar et al. (2021), takes a point subtracted input image and uses the Markov Chain Monte Carlo (MCMC) algorithm to run through the given halo parameters to automatically determine the most probable integrated flux and size. It also outputs a number of plots to visualize the MCMC run and final best-fitted halo. Could this package be incorporated into the CARACal pipeline to automatically measure halo flux?

I've attached the outputted log file from a run that recovers the manual flux measurement of a mini halo in A3444. A3444.DD.pt_sub.app.restored_mfs_deleted_axes_34.mini_halo_2d_cutout.added_crval3.fits_2022-01-29_18:41:04.log .

paoloserra commented 2 years ago

That'd be really interesting!

The best way to incorporate it into CARACal would be to package it into Stimela first -- https://github.com/ratt-ru/Stimela .

My suggestion would be to open an issue there, and as soon as the Stimela developers take care of it we could include Halo-CFDA into CARACal.

I guess this would need to be added at the very end of the selfcal worker?

KshitijT commented 2 years ago

Sounds great ! Maybe though the selfcal worker is now too big and this should be put in a separate worker?

viralp commented 2 years ago

Yes, it should go after the selfcal worker. However, one thing that needs to be considered here is that the Halo-FDCA works best for the point source subtracted image and hence simply using it for selfcal image may add bias in flux density measurements.

On Thu, Feb 3, 2022 at 11:32 AM Kshitij Thorat @.***> wrote:

Sounds great ! Maybe though the selfcal worker is now too big and this should be put in a separate worker?

— Reply to this email directly, view it on GitHub https://github.com/caracal-pipeline/caracal/issues/1406#issuecomment-1028783107, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABBBM4YRMIXGENFDHN2JOWDUZJDTXANCNFSM5NOKIKFQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you are subscribed to this thread.Message ID: @.***>

-- Postdoctoral Research Fellow South African Radio Astronomy Observatory (SARAO) and Rhodes University Square Kilometre Array, Cape Town South Africa

paoloserra commented 2 years ago

Just to clarify -- currently continuum imaging is done by the selfcal worker, independent of whether one wants to self-calibrate or not.

If I understand correctly Halo-FDCA is just an analysis tool, which should not be used while self-calibrating. It should just run on the final continuum image -- possibly following the subtraction of the point sources.

If so, I think that @KshitijT is right, and we would want this to be a separate "image analysis" worker, which could start with Halo-FDCA but may expand significantly in the future.

Finally, if Halo-FDCA needs an image with no point sources we would need to put some effort in the point source subtraction step.

ktrehaeven commented 2 years ago

I agree with @paoloserra. This should be in a future "image analysis" worker. And we still need to work on automating the point source subtraction. We can do it fairly easily with stimela and wsclean/ddfacet for 2gc data. I've been working on manually getting 3gc point source subtracted images but the automation of which is probably beyond me.

I'll open an issue on the stimela page regarding Halo-FDCA.

Athanaseus commented 2 years ago

I'll open an issue on the stimela page regarding Halo-FDCA.

Now available in stimela master.

I'm renaming this issue to Add image-analysis-worker.

This worker may include (open for discussion):