Download lots of Images

Carl Van Bogaert

unread,

Jul 27, 2021, 10:13:08 AM7/27/21

to astrometry

Hi Folks,

I'd like to download the largest archive of sky images that I can possibly find.

The plan is to train an artificial neural network to solve the field using the solved images as examples.

I don't much care if they're already solved or not - I can run the solver locally if needed. I just require A LOT of potentially solvable images.

Is there a way I can pull down nova.astrometry.net's archive of images without some silly web scraping? Anyone know of a larger publicly available archive?

Thanks,

Carl

Dustin Lang

unread,

Jul 27, 2021, 11:04:20 AM7/27/21

to Carl Van Bogaert, astrometry

You should be asking about the licensing terms.

--
You received this message because you are subscribed to the Google Groups "astrometry" group.
To unsubscribe from this group and stop receiving emails from it, send an email to astrometry+...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/afc19057-870a-4022-896b-eb5fbf017af4n%40googlegroups.com.

Dustin Lang

unread,

Jul 27, 2021, 11:14:06 AM7/27/21

to Carl Van Bogaert, astrometry

Again, being careful with licensing, you could check out astrobin.com (which also uses astrometry.net to tag astrometry).

cheers,

--dustin

On Tue, Jul 27, 2021 at 10:13 AM Carl Van Bogaert <cvanb...@gmail.com> wrote:

--

Dustin Lang

unread,

Jul 27, 2021, 11:15:21 AM7/27/21

to Carl Van Bogaert, astrometry

And of course there are professional sky surveys with excellent astrometry, including sdss.org and legacysurvey.org

John Murrell

unread,

Jul 30, 2021, 4:31:36 AM7/30/21

to Dustin Lang, Carl Van Bogaert, astrometry

The problem with the using the sky surveys are that each telescope only has one fixed image scale so you may well end up with a Neural Network that will only solve images of that particular scale.

If the intent is to solve random images you need a training set with a random scale and also all the problems with images that Astrometry.net has to cope with such as distortion, orientation, local lights etc.

John MUrrell

Wayne Green

unread,

Jul 30, 2021, 3:50:42 PM7/30/21

to John Murrell, Dustin Lang, Carl Van Bogaert, astrometry

IPAC and LCO have public repositories of images from lots of places. The LCO

scopes, while the 'same' specs are slightly different OTAs with varying PSFs

depending on site and engineering cycles.

--W

--

You received this message because you are subscribed to the Google Groups "astrometry" group.
To unsubscribe from this group and stop receiving emails from it, send an email to astrometry+...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/bec11a2fef8185087bd186e1e5977142c446cff3.camel%40JohnMurrell.org.uk.