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

Counting stars

  • Please log in to reply
11 replies to this topic

#1 Ns58

Ns58

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 02 Dec 2011
  • Loc: Vancouver BC/Black Butte Oregon

Posted 06 November 2018 - 11:08 AM

Hi is there software available to count stars, local peaks, in widefield images that would be accessible? I was just looking at some astroimages and would love to know how many stars I am resolvingo ther than just making an estimate.
Cheers
Joe

#2 TOMDEY

TOMDEY

    Surveyor 1

  • -----
  • Posts: 1862
  • Joined: 10 Feb 2014
  • Loc: Springwater, NY

Posted 06 November 2018 - 12:53 PM

A friend developed an algorithm for me, for counting scratches and digs on dark-field illuminated optical surfaces. Which is exactly the same challenge. The output would give binned totals for each, ranked according to size. He wrote code in a couple of days, so it can't be too crazy challenging. But, he is mathematician, etc.

 

With your star-counting thing... you would have to calibrate the output... probably according to magnitude. And your lens would have to be superbly free of ALL aberrations! The calibration part would BY FAR be the toughest part!  Tom

 

Attached Thumbnails

  • 054 counting algorithm.jpg


#3 Jaimo!

Jaimo!

    Mercury-Atlas

  • *****
  • Moderators
  • Posts: 2720
  • Joined: 11 Oct 2007
  • Loc: Exit 135 / 40° North

Posted 06 November 2018 - 11:01 PM

I do a lot of microscopy for my career, there is the free biology image analysis program 'ImageJ' with a multitude of plugins.  There should be an plugin specifically for counting ROI, or regions of interest.  With a little manipulation if your image you should easily come up with an acceptable method for counting.

 

YMMV,

Jaimo!



#4 robin_astro

robin_astro

    Viking 1

  • -----
  • Posts: 554
  • Joined: 18 Dec 2005

Posted 07 November 2018 - 06:03 PM

You could submit your image to nova.astrometry.net which will calibrate your image and generate a list of all the stars with their coordinates and names for the ones identified 

 

Robin


Edited by robin_astro, 07 November 2018 - 06:06 PM.


#5 Ns58

Ns58

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 02 Dec 2011
  • Loc: Vancouver BC/Black Butte Oregon

Posted 07 November 2018 - 11:57 PM

Robin I think this is for just identifying a specific region in the sky and some main objects? I want to actually count the number of detected stars in an individual star field that I am photographing.

 

I have attached a link to photos to give you an idea of what I really mean. A single exposure of the star field around the North American Nebula. Three versions of the same image but manipulated in Lightroom/photoshop. I simply think it would be fun to know how many stars I can detect in a single image using specific equipment under specific skies.  Detecting each luminosity peak will take some calculating. Jaimo's suggestion of ImageJ which looks interesting.

 

https://adobe.ly/2Dbh1Tk



#6 robin_astro

robin_astro

    Viking 1

  • -----
  • Posts: 554
  • Joined: 18 Dec 2005

Posted 08 November 2018 - 05:43 AM

Actually, astrometry.net produces a file (a fits table) of  all objects it detects as stars (even ones it cannot identify) which includes their location in the image and the flux. For example I just uploaded a random wide field    image of a region including M31 I found on the net and it found 3857 stars

 

Robin


Edited by robin_astro, 08 November 2018 - 05:45 AM.

  • Jaimo! likes this

#7 Ns58

Ns58

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 02 Dec 2011
  • Loc: Vancouver BC/Black Butte Oregon

Posted 08 November 2018 - 10:42 AM

Great thanks. I will try that.

Joe



#8 goodricke1

goodricke1

    Messenger

  • -----
  • Posts: 497
  • Joined: 18 May 2013
  • Loc: Ireland

Posted 08 November 2018 - 01:26 PM

Actually, astrometry.net produces a file (a fits table) of  all objects it detects as stars (even ones it cannot identify) which includes their location in the image and the flux. For example I just uploaded a random wide field    image of a region including M31 I found on the net and it found 3857 stars

 

Robin

 

Can you advise as to which fits file viewer you used to get that result? Thanks.



#9 Ns58

Ns58

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 02 Dec 2011
  • Loc: Vancouver BC/Black Butte Oregon

Posted 08 November 2018 - 03:13 PM

How sis you access the output file? I did not see where there is one with a test.

Joe



#10 Ns58

Ns58

    Explorer 1

  • -----
  • topic starter
  • Posts: 83
  • Joined: 02 Dec 2011
  • Loc: Vancouver BC/Black Butte Oregon

Posted 08 November 2018 - 03:24 PM

Here is an attempt using my file and astronomy.net

 

Go to results page
Job 3030694:
Failed

Source extraction image (fullsize)
Log file tail [-]
(full)
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
After removing quad stars: no reference stars
object 37 of 37: 4090320 quads tried, 5927280 matched.
Field 1: tried 4090320 quads, matched 5927280 codes.
Field 1 did not solve (index index-4107.fits).
Best odds encountered: 459778
Spent 568.888 s user, 2.856 s system, 571.744 s total, 572.324 s wall time.
cx<=dx constraints: 0
meanx constraints: 19243885
RA,Dec constraints: 0
AB scale constraints: 0
Spent 572.325 seconds on this field.
Log file 2



#11 robin_astro

robin_astro

    Viking 1

  • -----
  • Posts: 554
  • Joined: 18 Dec 2005

Posted 09 November 2018 - 11:20 AM

Can you advise as to which fits file viewer you used to get that result? Thanks.

You need one which can manage tables. I used fv

https://heasarc.gsfc....gov/ftools/fv/

 

It looks though that there are just too many stars in this image for astrometry.net to handle

 

A quick and dirty way to estimate the number of stars would be to randomly select some small areas and count them by hand and scale up. The variation between the individual areas would give you a estimate of your uncertainty in the total.

 

Cheers

Robin



#12 goodricke1

goodricke1

    Messenger

  • -----
  • Posts: 497
  • Joined: 18 May 2013
  • Loc: Ireland

Posted 09 November 2018 - 01:15 PM

You need one which can manage tables. I used fv

https://heasarc.gsfc....gov/ftools/fv/

 

Yes that works great, cheers!




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