problem running annotations

[Astro Keith]

When running the solver on my Pi (on an image 'capture.jpg'), the output file capture-nbc.png contains no annotations. If I upload it it comes back annotated (has M13 in the middle for instance).

When I try and run plotann.py separately, I get the following error

anwcs.c:917:getheader: Failed to open file capture.new
anwcs.c:1061:anwcs_open_wcstools: WCStools support was not compiled in
plotstuff.c:421:plotstuff_set_wcs_file: Failed to read WCS file "capture.new", extension 0
Segmentation fault

I'm assuming capture.new is the wcs.fits file required.

Is this an error with the file?

Or is this an error with my astrometry.net build?

As always - help much appreciated

Keith

[Dustin Lang]

capture.new is your original image converted to FITS format with a WCS header added.  There should also be a capture.wcs file containing just the WCS header that you can pass to plotann.py .

cheers,

dustin

--
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/017db4ef-b5d6-4f1d-badd-c352657f6f4dn%40googlegroups.com.

[Astro Keith]

Thanks.

Should have said that I tried capture.wcs. -  get the same error message

Keith

[Astro Keith]

Sorry - had a path error - getting plots now.

Thanks!

On Wednesday, June 29, 2022 at 2:56:58 PM UTC+1 dstn...@gmail.com wrote:

[Astro Keith]

New issue!

I'm getting grid lines but nought else! Suspect its because I don't have any catalogues installed. I've searched my folder system for all *.fits files and no catalogues were found.

My astrometry.net install was courtesy of Astroberry, and I'm thinking its not complete?

Can you point me to a download source for the catalogues please?

Thanks

Keith

[Dustin Lang]

Depending on the scale of your image there won't always be objects to annotate.  The built-in annotator knows about NGC/IC objects (compiled in) and a few dozen bright stars.

cheers,

dustin

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/156f96ba-fee0-4efe-bdf8-3419068e0d9bn%40googlegroups.com.

[Astro Keith]

Thanks,

I found and downloaded hd.fits and hip.fits at data.astrometry.net . Invoking these produces overlays.

Couldn't find or download any other catalogues referenced in the --help.

My image of M13 is 80' * 60' . As well as M13 which is visible, should have many ngc's not resolved, but none are plotted. Are only those detected in the image plotted?

I'm following the -verbose options and at the moment haven't included any specific options to see ngc objects as the options seem to be about turning them off!

As I said, if I use the on-line solver on the same image - it comes back with many ngc annotations.

I tried another image which included Polaris and indeed that is annotated regardless of options. Excellent.

So I'm making progress (I think!)

Thanks  again

Keith

[Dustin Lang]

Hi,

For plotann.py there's also the tycho2.kd file on data.astrometry.net,

and uzc2000.fits, abell-all.fits, and ugc.fits distributed with the code in https://github.com/dstndstn/astrometry.net/tree/main/catalogs

There was a time when the Debian Astrometry.net package omitted the NGC/IC catalog we were using because of licensing issues.  Maybe you have a version derived from that.  In current versions we're using a Debian-approved NGC/IC catalog.

cheers,

dustin

To view this discussion on the web visit https://groups.google.com/d/msgid/astrometry/adf2b614-c10c-4b38-bc01-80c5c5bcc3e5n%40googlegroups.com.

[Dustin Lang]

Hi,

(replying back to the list)

If you want to send exactly what you're running and the error messages you're getting, I may be able to diagnose what's going on.  By default, it installs into /usr/local/astrometry, so the executables will not be on your path.  You would have to

export PATH=${PATH}:/usr/local/astrometry/bin

to get it to work.  Alternatively, you can install it in a more standard place (which does make it harder to remove...), by

make install INSTALL_DIR=/usr/local

plotann.py is in plot/plotann.py and should get installed into

bin/plotann.py (as a symlink to lib/python/astrometry/plot/plotann.py)

cheers,

dustin

On Fri, Jul 1, 2022 at 6:42 AM Keith Venables <k.ven...@ntlworld.com> wrote:

Many thanks,

I tried the latest version of astroberry - still not there.

I am now building a full copy of astrometry.net on a fresh Debian10 Pi OS.

Just for your info I came across some issues following the install notes at http://astrometry.net/doc/build.html probably they are a bit old now?

Install python-numpy, python-pyfits reports as not found.

pip install numpy and pip install pyfits work.

Install python-dev says it wants to remove the Python-is-python3 and change to python 2.7. I declined - was this correct on my part?

Install zlib1g-dev is displayed on my browser (Safari) as zliblg-dev, ie the 1 is shown as an ‘l’. Googleing for help revealed a lot of others have had the same issue!

Copy and paste worked fine.

The steps to build, all needed to be run as sudo on my Pi.

I’m now working on whatever I need to do to configure. Although I’ve added a PATH to the binaries, my solve commands are being rejected as not found.

I’ve also noticed that plotann.py which I was using previously, doesnt seem to be there any more? I see plotstuff.py. Has it been deprecated?

Best regards

Keith

On 30 Jun 2022, at 17:17, Dustin Lang <dstn...@gmail.com> wrote:

plotann.py uses the C code to plot the NGCs -- so it's using the same compiled-in NGC list (empty, in your case) as solve-field's plots.

On Thu, Jun 30, 2022 at 11:59 AM Keith Venables <k.ven...@ntlworld.com> wrote:

Hi Dustin

I’m still not getting any ngc objects plot for labelled.

I can now get HIP, HD & Tycho2 overlays fine. Thanks.

I’ve manually downloaded openngc-names.fits, openngc-ngc.fits, openngc-ic.fits files and saved them with permissions. 

This is the script I am using at the moment

<Screenshot 2022-06-30 at 16.48.56.png>

This is the output

<Screenshot 2022-06-30 at 16.53.56.png>

If I upload the same image for a solve, I get returned

<Screenshot 2022-06-30 at 16.46.58.png>

The plotann.py version on my Pi is the same as on the GitHub.

The verbose output says "Checking 0 NGC/IC objects"

 <Screenshot 2022-06-30 at 16.49.25.png>

Any more ideas you can think of!

Regards

Keith