problems with index range?

Mariana Rubet

unread,

Nov 16, 2021, 8:34:54 AM11/16/21

to astrometry

Hi Dustin,
I'm running Astrometry on my notebook (thank u for the support, it helped me a lot!) and there are some images that Astrometry couldn't set the wcs (but it did on the website). Should I download indexes 4200-*.fits and 4201-*.fits as well? My region is 23 arcmin in diameter. I've configured the astrometry.cfg file with the indices 4202-4211*.fits .

Dustin Lang

unread,

Nov 16, 2021, 8:51:04 AM11/16/21

to Mariana Rubet, astrometry

Hi,

With images 23 arcmin across, you should not need scales 00 and 01; those have quadrangles of stars 2 to 4 arcmin across, so they will be small in your images. Scales 03-04 are probably the most useful. Which index file did the web version find the answer in? If you go to to Submission status page, toward the end of the log file you should see a line like

"Field 1: solved with index index-207.fits."

If the 4200 series doesn't work, I would try the 5200 series. Also make sure the source detection looks okay (look at the *-objs.png plots).

cheers,

--dustin

On Tue, Nov 16, 2021 at 8:34 AM Mariana Rubet <marian...@astro.ufrj.br> wrote:

Hi Dustin,
I'm running Astrometry on my notebook (thank u for the support, it helped me a lot!) and there are some images that Astrometry couldn't set the wcs (but it did on the website). Should I download indexes 4200-*.fits and 4201-*.fits as well? My region is 23 arcmin in diameter. I've configured the astrometry.cfg file with the indices 4202-4211*.fits .

--
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/bab24d91-c5ea-4c39-aa8a-738da4fb7415n%40googlegroups.com.

Mariana Rubet

unread,

Nov 16, 2021, 10:24:33 AM11/16/21

to astrometry

The web version : "Field 1: solved with index index-202-06.fits"

Isn't it the same of index 4202-06?

Dustin Lang

unread,

Nov 16, 2021, 11:03:24 AM11/16/21

to Mariana Rubet, astrometry

You could also try index 5206. You would want healpixes healpixes 24,25,26,27:

https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/LITE/index-5206-24.fits

https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/LITE/index-5206-25.fits

https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/LITE/index-5206-26.fits

https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/LITE/index-5206-27.fits

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/2e0355e7-efef-4caa-b8a6-7fbdbf860788n%40googlegroups.com.

Mariana Rubet

unread,

Nov 16, 2021, 5:56:59 PM11/16/21

to astrometry

Not solving yet with indexes 5206-24/25/2/27. . Should I try with the indexes 5202.-*fits ?

Dustin Lang

unread,

Nov 16, 2021, 6:59:37 PM11/16/21

to Mariana Rubet, astrometry

I'm sorry, I meant 5202! 5202-24, 25, 26, 27. Sorry about that!

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/72e1e564-2dcb-4ef2-8317-b1cf305745bcn%40googlegroups.com.

Mariana Rubet

unread,

Nov 18, 2021, 10:03:37 AM11/18/21

to astrometry

Hi Dustin,

I think I will have to download more healpixeis from index 5200 (couldn't solve wcs yet, for some images). I tried to change the wget.sh file from index 4200 but its not working. Is the "build.sh" file (in the index-5200 folder https://portal.nersc.gov/project/cosmo/temp/dstn/index-5200/) used for that? I Thought to download all healpixels 5202- *

Dustin Lang

unread,

Nov 18, 2021, 10:11:22 AM11/18/21

to Mariana Rubet, astrometry

Hi,

You can generate the wget commands like this:

cheers,

--dustin

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/fd753132-dfd5-4cfa-9d35-993a2b833be8n%40googlegroups.com.

Mariana Rubet

unread,

Nov 19, 2021, 9:19:15 AM11/19/21

to astrometry

Hi Dustin,

Even with all the healpixeis 5202- * it's not solving the wcs \o/

Here goes de screenshot of the Astrometry website for this image

I have ~400 images with no wcs. For many of them, Astrometry has succeeded solving wcs. That's why I think the problem must be an index issue.

Screenshot from 2021-11-19 10-58-28.png

Dustin Lang

unread,

Nov 19, 2021, 9:29:27 AM11/19/21

to Mariana Rubet, astrometry

Hi Mariana,

What solve-field command are you using?

Do the *-objs.png plots look good (are stars correctly detected & centered?)?

(The web version does "--downsample 2" by default.)

cheers,

--dustin

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/fde6620e-5e69-4fd1-9de9-2136e34e9874n%40googlegroups.com.

Dustin Lang

unread,

Nov 19, 2021, 9:40:16 AM11/19/21

to Mariana Rubet, astrometry

Hi Mariana,

Strange, I just tried that example image and I do get a solution in index-5202-25.fits, after looking at object 21.

To speed things up, I did:

solve-field /tmp/m.fits --config /tmp/cfg --scale-low 0.18 --scale-high 0.22 --scale-units app --ra 150 --dec 3 --radius 5 -v --continue --plot-scale 0.25 --no-tweak

cheers,

--dustin

Mariana Rubet

unread,

Nov 19, 2021, 3:26:25 PM11/19/21

to Dustin Lang, astrometry

It's solving with this command line you sent to me (and very fast) =) Thank u!

Reply all

Reply to author

Forward