Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Plate solve and label your images with PixInsight

  • Please log in to reply
19 replies to this topic

#1 G. Hatfield

G. Hatfield

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1572
  • Joined: 31 Jan 2009
  • Loc: Tucson, Arizona

Posted 26 March 2015 - 01:54 PM

Forgive me if you already know about this feature of PixInsight, but I was unaware of it.   These two scripts can plate solve and label your images.  

 

ImageSolver (under image analysis)

Annotateimage (under render)

 

And there is even some directions to go with the plate solver.  You must give an approximate RA and DEC (it can't do it blind) and when it is done, it writes the info to the image file.  Then use the annotater to label the image.  Pretty amazing.  See the attached examples.  In the following post, I show an example of the labeled image. 

Attached Thumbnails

  • image plate solver.jpg
  • image annotation.jpg

Edited by G. Hatfield, 26 March 2015 - 02:01 PM.

  • Tim, artem2, jakejake and 2 others like this

#2 G. Hatfield

G. Hatfield

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1572
  • Joined: 31 Jan 2009
  • Loc: Tucson, Arizona

Posted 26 March 2015 - 01:54 PM

Here is the labeled image of the Leo Triplet.

 

George

 

P.S.  I should mention that I heard about this on the Yahoo Group....DSLR Astro Image Processing... thanks Mike and Alex!

Attached Thumbnails

  • labelled image.jpg

Edited by G. Hatfield, 26 March 2015 - 02:10 PM.

  • guyroch and dhammy like this

#3 nitesky

nitesky

    Vostok 1

  • -----
  • Posts: 176
  • Joined: 28 Mar 2008
  • Loc: North Carolina, USA

Posted 26 March 2015 - 02:35 PM

Holy cow!!!! I've been looking for something like that. I hadn't seen that feature.  MANY THANKS, George!



#4 HxPI

HxPI

    Apollo

  • -----
  • Posts: 1328
  • Joined: 05 Sep 2013
  • Loc: Virginia Beach, VA

Posted 26 March 2015 - 10:18 PM

This is awesome! I just tried it out and it worked great. Thanks for sharing.

 

Ciao,

Mel



#5 jwheel

jwheel

    Surveyor 1

  • *****
  • Posts: 1683
  • Joined: 23 Jan 2008
  • Loc: Fort Davis TX

Posted 27 March 2015 - 10:36 AM

Thanks!



#6 whwang

whwang

    Vanguard

  • *****
  • Posts: 2489
  • Joined: 20 Mar 2013

Posted 27 March 2015 - 11:05 AM

Thanks for sharing.  I was not aware of these functions either.

 

Cheers,

Wei-Hao



#7 Greenbbs

Greenbbs

    Ranger 4

  • *****
  • Posts: 306
  • Joined: 21 Oct 2014
  • Loc: Victor, NY

Posted 27 March 2015 - 02:15 PM

I tried it today and it worked!!!! Now I'll just save a non solved and a solved copy!!

#8 Wigleydh

Wigleydh

    Vostok 1

  • -----
  • Posts: 144
  • Joined: 26 Oct 2007

Posted 28 March 2015 - 08:56 AM

I've used this a couple of times and it does work good.



#9 nitesky

nitesky

    Vostok 1

  • -----
  • Posts: 176
  • Joined: 28 Mar 2008
  • Loc: North Carolina, USA

Posted 28 March 2015 - 02:54 PM

2015_M106_annotated_sm.jpg

 

 

Thanks George for your assistance. Your information got me over the hump and I discovered what I was doing wrong.

 

I love this!!

 

 



#10 G. Hatfield

G. Hatfield

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1572
  • Joined: 31 Jan 2009
  • Loc: Tucson, Arizona

Posted 28 March 2015 - 03:15 PM

I've learned a couple of things.  

 

First, I was having trouble getting the plate solve script to work on some of my images.  I would put in the focal length of my scope and the pixel size of the camera, but it would "blow up" and not solve the image.  Then it occurred to me that the images I was using had been reduced in size. When I put in a corrected image scale (i.e., corrected for the fact that I had reduced the image size by 2/3, from 5094 x 3414 pixels to 1728 x 1158 pixels) it worked on every image.  So my "native" image scale, which is about 1.38 arcsec/pixel, had to be entered as 4 arcsec/pixel for the resized (not cropped) image.  

 

Also the search function works very well.  I was looking up the RA and DEC in SkyX, but the search function will find these values for most objects even when the common name is used.

 

Sometimes the labeling from the Tycho-2 catalog can overwhelm the image.  If you highlight this catalog a filter can be applied to limit the stars to a particular mag range.  

 

George


Edited by G. Hatfield, 28 March 2015 - 03:19 PM.

  • HxPI likes this

#11 HxPI

HxPI

    Apollo

  • -----
  • Posts: 1328
  • Joined: 05 Sep 2013
  • Loc: Virginia Beach, VA

Posted 28 March 2015 - 03:30 PM

I've learned a couple of things.  

 

First, I was having trouble getting the plate solve script to work on some of my images.  I would put in the focal length of my scope and the pixel size of the camera, but it would "blow up" and not solve the image.  Then it occurred to me that the images I was using had been reduced in size. When I put in a corrected image scale (i.e., corrected for the fact that I had reduced the image size by 2/3, from 5094 x 3414 pixels to 1728 x 1158 pixels) it worked on every image.  So my "native" image scale, which is about 1.38 arcsec/pixel, had to be entered as 4 arcsec/pixel for the resized (not cropped) image.  

 

Also the search function works very well.  I was looking up the RA and DEC in SkyX, but the search function will find these values for most objects even when the common name is used.

 

Sometimes the labeling from the Tycho-2 catalog can overwhelm the image.  If you highlight this catalog a filter can be applied to limit the stars to a particular mag range.  

 

George

 

Today, while working on the Orion Nebula, I also realized the same issue of pixel size for an image that was cropped/enlarged. It's pretty finicky with the tolerance of the input! I suppose it forces better cataloging of image attributes. Like they say, garbage in garbage out!

 

Thanks for sharing.

 

Ciao,

Ml


Edited by HxPI, 28 March 2015 - 03:31 PM.


#12 G. Hatfield

G. Hatfield

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1572
  • Joined: 31 Jan 2009
  • Loc: Tucson, Arizona

Posted 16 April 2015 - 07:38 AM

I recently learned another critical factor in setting this up.  The limit magnitude must be set to about 18 for it to work in some instances.  In fact, if you do that and set the RA and DEC correctly it will often work with everything else with defaults.

 

George



#13 Tonk

Tonk

    Cosmos

  • *****
  • Posts: 8800
  • Joined: 19 Aug 2004
  • Loc: Leeds, UK, 54N

Posted 16 April 2015 - 08:09 AM

So Astrometry.net knows how to plate solve **without** RA/DEC and FOV/Scale hints (Blind Solver). So why does PI require these hints and not blind solve also?

 

You have the bizarre situation if you pick up an old image that you have forgotten where it was in the sky and want to annotate it you have send if off to Astrometry.net for analysis to be able to tell PI where it actually is and at what scale? 

 

I think the Astrobin guys did an excellent job piggy-backing on Astrometry.net web services to support their annotation tool. I wonder if you can call up Astrometry.net via PI in a similar manner to seed the location/scale info into the PI annotator?? Anyone know?


Edited by Tonk, 16 April 2015 - 08:18 AM.


#14 coinboy1

coinboy1

    Apollo

  • *****
  • Posts: 1123
  • Joined: 03 May 2011
  • Loc: Tulsa, OK

Posted 16 April 2015 - 05:58 PM

This is awesome. I just used it and it works perfectly!!



#15 powerstroke01

powerstroke01

    Viking 1

  • *****
  • Posts: 605
  • Joined: 19 Jun 2007
  • Loc: Western Sierra Nevada Foothills

Posted 17 April 2015 - 11:37 AM

Now i want PI!!!



#16 Orion58

Orion58

    Apollo

  • -----
  • Posts: 1161
  • Joined: 28 Oct 2011
  • Loc: Southern Wisconsin

Posted 04 December 2015 - 01:21 PM

OK, it took me a while to see this thread, but these PI features are VERY cool...Thanks for sharing this info George!



#17 bmhjr

bmhjr

    Viking 1

  • ****-
  • Posts: 984
  • Joined: 02 Oct 2015
  • Loc: Colorado

Posted 04 December 2015 - 03:20 PM

So Astrometry.net knows how to plate solve **without** RA/DEC and FOV/Scale hints (Blind Solver). So why does PI require these hints and not blind solve also?

 

You have the bizarre situation if you pick up an old image that you have forgotten where it was in the sky and want to annotate it you have send if off to Astrometry.net for analysis to be able to tell PI where it actually is and at what scale? 

 

I think the Astrobin guys did an excellent job piggy-backing on Astrometry.net web services to support their annotation tool. I wonder if you can call up Astrometry.net via PI in a similar manner to seed the location/scale info into the PI annotator?? Anyone know?

 

You can upoad an image to Astrometry.net and then download the plate solved .fts file that has the headers embedded.  Feed that to PI Annotate script.


Edited by bmhjr, 04 December 2015 - 03:20 PM.

  • artem2 and jakejake like this

#18 iwols

iwols

    Ranger 4

  • -----
  • Posts: 354
  • Joined: 27 Jul 2016
  • Loc: doncaster uk

Posted 11 August 2016 - 08:32 AM

Forgive me if you already know about this feature of PixInsight, but I was unaware of it.   These two scripts can plate solve and label your images.  

 

ImageSolver (under image analysis)

Annotateimage (under render)

 

And there is even some directions to go with the plate solver.  You must give an approximate RA and DEC (it can't do it blind) and when it is done, it writes the info to the image file.  Then use the annotater to label the image.  Pretty amazing.  See the attached examples.  In the following post, I show an example of the labeled image. 

great can you give me a little more info how to use looked all over pixinsight  but cant find anything cheers



#19 Daniel Dance

Daniel Dance

    Eastern Voltage Research, LLC

  • *****
  • Vendors
  • Posts: 1651
  • Joined: 08 Jun 2016
  • Loc: Philadelphia, PA

Posted 11 August 2016 - 11:32 AM

I bet photoshop can't do that!\



#20 G. Hatfield

G. Hatfield

    Surveyor 1

  • *****
  • topic starter
  • Posts: 1572
  • Joined: 31 Jan 2009
  • Loc: Tucson, Arizona

Posted 11 August 2016 - 11:55 AM

To label your image with PI you need to use two scripts.  First open your image in PI.

 

1.  Under image analysis, use imagesolver.  Three things to note.  First, PI needs to know what portion of the sky you want labeled.  Select search and put int he NDC, IC or star name for the field in question.  It will usually be able to find it, but not always.  Or you can just put in the RA and DEC. Second, the program needs to know the image scale you are using.  This can be calculated if the program has the focal length in mm and the pixel size for your camera.  Third, if you have a problem getting it to work, try a different star catalog.  I usually set the limiting magnitude to 18.  Once you have a solution, then go to step 2.

 

2.  Under render, use annotateimage.   You will probably want to limit the number of stars labeled if you are using Tycho-2.  Highlight this catalog to get to its configuration dialog box.  Set the minimum magnitude to 8 or so (right box).  If you get overlapping labels, you can define what side of the star or object is labeled..... so put one label on one side and the other on the other side.

 

George


  • artem2 and iwols like this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics