Star Hopping with on-the-fly Astrometry

[badgers]

Yes the wcs-xy2rd is buried inside a subroutine in there somewhere.

I run two copies of this script now, one on ports 10010 and 10015 for solving the current image and I keep another one running at 10030 and 10040 which shows the target image.

No new progress on stellarium hacks, work has taken over my free time : (

[badgers]

I recently bought a netbook and managed to get this up and running in windows 7, so I can assist others who want to try it on windows.

The Cygwin install that comes with Astrotortilla is extremely useful and gets you 90% of the way there.

The only issue I'm seeing is that solves and also image conversions are very very slow on the intel atom processor on the netbook.

Stellarium connections, solves, corners and everything else seem to work just peachy though.

Anton

[aruangra]

Hi Anton,

I came across your interesting post. I am going to integrate Cartes du Ciel with Astrometry.net program to show solved fields. I don't know how to get the orientation from Astrometry.net. I get only the field size and center RA&DEC. Do you have any suggestion? CdC can show a CCD finder and accepts command line inputs. So, I will interrupt CdC somewhere to show the solution from Astrometry.net.

Best regards,

Anat

[themos]

Anat, the entire image is characterised by the .wcs file that astrometry.net produces. Use the wcsinfo program to print out all the fields nicely. In particular, there is an "orientation" field. Here's a printout of the output of wcsinfo on the demo1 image.

• crpix0 398.0027324
  
• crpix1 264.881163037
  
• crval0 100.21316225
  
• crval1 9.8267875768
  
• ra_tangent 100.21316225
  
• dec_tangent 9.8267875768
  
• pixx_tangent 398.0027324
  
• pixy_tangent 264.881163037
  
• imagew 800
  
• imageh 526
  
• cd11 -3.64287634481e-05
  
• cd12 -0.00188127996703
  
• cd21 0.00188234437892
  
• cd22 -3.67727511807e-05
  
• det 3.54255635697e-06
  
• parity 1
  
• pixscale 6.77580477776
  
• orientation -91.114248
  
• ra_center 100.215707015
  
• dec_center 9.83153907822
  
• orientation_center -91.113813
  
• ra_center_h 6
  
• ra_center_m 40
  
• ra_center_s 51.7696836306
  
• dec_center_sign 1
  
• dec_center_d 9
  
• dec_center_m 49
  
• dec_center_s 53.5406815868
  
• ra_center_hms 06:40:51.770
  
• dec_center_dms +09:49:53.541
  
• ra_center_merc 0.27837696
  
• dec_center_merc 0.52744484
  
• fieldarea 1.49071
  
• fieldw 90.36
  
• fieldh 59.38
  
• fieldunits arcminutes
  
• decmin 9.06879
  
• decmax 10.5939
  
• ramin 99.6976
  
• ramax 100.731
  
• ra_min_merc 0.27981
  
• ra_max_merc 0.276938
  
• dec_min_merc 0.529597
  
• dec_max_merc 0.525297
  
• merc_diff 0.00429957
  
• merczoom 8
  

[aruangra]

Thanks so much

Anat

[badgers]

Looking forward to seeing the results of your CdC approach Anat, it sounds very nice.

Also I've migrated to using sextractor to find stars in the image prior to running solve-field and am also using the FITS header information (image scale or approximate RA/DEC) to get solves on my slow netbook happening in 3-5 seconds.

Anyone interested in using sextractor be aware that compiling it is a nightmare as it requires many other dependencies which are not always trivial to get hold of.

Anton

[aruangra]

It is amazing that I can get blind-solved results in 5 sec on MK808. I skip the modification to show the CCD frame for now. So I can work on the integration of plate-solving and CdC first which is pretty straightforward because CdC accepts server commands via a TCP socket:

http://www.ap-i.net/skychart/en/documentation/server_commands#v27_compatibility_dde_command

[themos]

:hello2:

[simonct]

Hi All,

I've been following this thread for a little while and thought I'd finally drop in.

If you're on a Mac, I've put together a simple app that embeds all the astrometry.net binaries and dependent libraries into a single download, so no need to deal with package managers or compiling from source.

It's pretty basic, but it should let you try out the solver without building it all yourself.

You'll need 10.7+ (I can spend some time getting it running on 10.6 if anyone wants) and to supply your own indexes.

To use it just launch the app, choose the location of the indexes and then drop an image file onto the window and press Solve. It currently only works on standard image files - no fits or raw as yet.

I've only tried it on two machines so there's a chance it may not work on yours - any and all bug reports and feature requests welcome.

You can get the app here

Enjoy !

Simon

[aruangra]

I finally compiled CdC on MK808 and use CdC via server commands. This is the plate-solved image overlaid on the skychart

The astrometry and CdC is interfaced via a web application.

[Gonzo]

Anton,

I'm trying your script (latest one), I think I'm missing something as there's nothing much happening in Stellarium (0.11.1).

I believe I have set the connection correctly as I do see your script saying: connected to stellarium.

Could you please let me know how you've setup stellarium?

[simonct]

I've uploaded an update to my Mac app. The main change is that it now renders the annotations itself rather than using the bitmap generated by the solver which means you can now select which objects to show, choose the colour and font and the annotations stay nice and smooth independent of zoom level. You can also save the annotated image to a file.

You can get it here.

Simon

[Gonzo]

Looking forward to seeing the results of your CdC approach Anat, it sounds very nice.

Also I've migrated to using sextractor to find stars in the image prior to running solve-field and am also using the FITS header information (image scale or approximate RA/DEC) to get solves on my slow netbook happening in 3-5 seconds.

Anyone interested in using sextractor be aware that compiling it is a nightmare as it requires many other dependencies which are not always trivial to get hold of.

Anton

I agree, compiling sextractor is a serious pain the back side!!!!

[Epicycle]

Hi Simon,

just to let you know that I've installed your app in my mac and works perfectly, simple and straight to the point. And it's much easier than installing astrometry as instructed in astrometry.net.

Eventually one hopes that there will be an astrotortilla for mac via EqMac, but in the meantime your app with skysafari (which I prefer to stellarium) does the trick for me. Thanks!

E.

[simonct]

Thanks - glad you're finding it useful !

I've talked to OzDave about integrating with EQMac and will post an update if I get something working.

Simon

[Gonzo]

Hi Anton,

Any news on how to use sextractor please?

[Gonzo]

Just tried again this script on a picture I took last night, it works perfectly fine. All the errors I was having the other day was down to me.

[Huskvarna]

Hello!

I found this forum and thread with google, because i am trying to get the solve information from Astrotortilla to show up in Stellarium.

And searching for "Astrotortilla stellarium port" gives this thread here on Stargazerslounge...

I read the thread and i am very fascinated...

What would i need to do to get it to work in Windows 7 with Astrotortilla and Stellarium installed?

Best Regards

Alexander

[Huskvarna]

Due to the overwhelming responses i recieved i decided to do a little trial and error...

And i actually got your script to work in cygwin, it solves the pictures i drop in a folder and creates a "_solved" jpg that's annotated!

I had to change some path's in the script since the astrotortilla package has astrometry installed in a different place. And install imagemagik. Dcraw did not exist in cygwin repositories so i had to install gcc to compile dcraw from source.

http://www.eecg.utoronto.ca/~aamodt/ece242/cygwin.html

http://www.cybercom.net/~dcoffin/dcraw/

Compile with "gcc -o dcraw -O4 dcraw.c -lm -DNODEPS"

Then the script said it was missing a font, so i fixed it with this:

http://blogs.bu.edu/mhirsch/2013/03/cygwin-imagematick-missing-usrsharefontsttfarial-ttf/comment-page-1/#comment-5722

Whew!

I'm quite happy with it working as it does now, would be nice if it could solve quicker tho... I have a Canon 1000d connected to a 150/750 newt, are there any settings that can be tweaked?

I will see if i can get it to work with stellarium also...

Thanks again for the script!

[uhb1966]

Hi everybody,

just wanted to drop in say say "Coool, Dudes...way out!"

I was thinking about using astrometry.net for plate solving, too... but you're way faster !! :grin:

[uhb1966]

I am now severely questioning myself whether i really need a goto mount

[uhb1966]

I've uploaded an update to my Mac app. The main change is that it now renders the annotations itself rather than using the bitmap generated by the solver which means you can now select which objects to show, choose the colour and font and the annotations stay nice and smooth independent of zoom level. You can also save the annotated image to a file.

You can get it here.

Simon

This runs like a treat! Takes about 8 secs on a 2010 Macbook Pro running OSX 10.9.4 (wide angle shot  with index 4100)

Fantastic!

[Bob Andrews]

Could it be that I missed it by 3 months?
Unfortunately simonct's link is dead. And he hasn't been online since september. :-(

Could anyone help me out with the app please? I've been fighting all these years to keep strictly on osx with my rig (eqmac, skysafari, phd, nebulosity, capture sequencer and even motor focus control) and learned to live with the throwbacks. But missing this chance for native plate solving feels like missing the sn in m82 last January…

[uhb1966]

No prob Bob, what do you need?! The app?

[NickK]

I'm thinking, after adding ATIK to OpenPHD (here), I'll probably look at how best to take resolving images.. it would be good to have something to resolve images for SGLX.