Star Hopping with on-the-fly Astrometry

[badgers]

Here's my updated script...

Any other fixes appreciated:

http://www.ebi.ac.uk/~aje/dist/starhop.pl

The way to run this for fast solving without plotting is:

perl starhop.pl --stellarium --noplot

For Solving and creation of solved images (quite a bit slower):

perl starhop.pl --stellarium

The new version with --noplot is much faster.

This version looks for .FITS, .FIT, .fits, .fit , .CR2 and .cr2

This version also prints out the orientation information in degrees, for some reason if mine outputs -60 degrees I need to ask for +60 degrees in stellurium for an accurate rotation.

Remember to edit the directory in the script that it analyses.

I've noticed that it can get locked in a cycle where it repeatedly fails to solve once a single image solve fails.

[Black Knight]

Yay! The Stellarium link-up is not too bad, try it!

What I get in Stellarium is a small cicle in the correct location (I think), labelled Astrometry Server. Is that correct? I don't get the larger circle and red-coloured rectangle shown in Anton's screen dump (on page 1 of this thread). Do you get that from the ocular plug-in, or is something not working quite right with starhop.pl?

[badgers]

Yes Black Knight, thats correct

The way I do it is this:

1) First make sure Stellarium oculars is set up for your scope and CCD as accurately as possible.

2) Start the starhop script and it will enable the little reticule "Astrometry Server", it defaults to polaris when no images are solved

3) I tend to click on the reticule or near it, Stellarium normally picks a star nearby and not the actual reticule.

4) Go to an ocular view (heavy zoom) and then click on the actual reticule so it's actually selected.

5) Now you can drop out of ocular view and go to the CCD view.

You should now have a red CCD box centred on the Astrometry reticule that will automatically pan with new solves.

Fiddly I know, but it works fairly well.

Anton

[Black Knight]

OK, I'll give that a go. Thanks Anton

[badgers]

No worries, I'm just thrilled to see people using this.

I'm also terrified people will see the sorry state of my programming and report me to the coding police.

Once in my life I wrote in C and actually knew what I was doing......

Anton

[badgers]

Actually, I've just discovered this....

Instead of faffing about trying to click the reticule to centre on it... Just click search in Stellarium and type "Astrometry Server" and it will do it in one go.

Also, debugging the version I just posted, i'll repost in a few mins.

I'm trying to have it work for jpg, fits and CR2 files and integrate the changes made by others...

I'm also going to add downsampling and image scale parameters so it's easier to set up for your system.

Anton

[badgers]

Ok, I've fixed a number of bugs.

1) The program will warn if dcraw or imagemagick are not available and will disable CR2 file or JPEG,FITS files accordingly

2) The program has two new options --downsample=X and --stars=Y to set the downsample rate (I use 3.0 and 500 respectively, these are defaults).

3) The lockup bug is fixed if an image is unsolved.

4) Rotation information seems to work just fine.

The updated version is here:

Please PM me back and changes to this code so I can keep an updated working version of the script.

http://www.ebi.ac.uk/~aje/dist/starhop.pl

Thanks,

Anton

[Black Knight]

I'll try that out later. I shouldn't worry about your coding, I've seen some pretty bad coding in my time ....

[themos]

Any chance of adding an environment variable that holds additional solve-field options? That way, those who want to tweak their solver can do it without modifying the script.

[themos]

I think the call to wcsinfo should be right after the print "SOLVED" line. If it didn't solve there won't be a .wcs file anyway.

[themos]

It might be a good idea to make your own temporary directory and clean up properly at the end. I find that some large files can be left behind by the script. Also, how do you end the script gracefully?

[badgers]

Points 1 and 2 are now done.

The program looks for an environment variable "SOLVEPARAMS" that will override any internal options.

This is passed to solve-field as follows:

solve-field /tmp/go.ppm SOLVEPARAMS

So the SOLVEPARAMS variable should contain a useful set of options, internally it is set to:

"-p -z $downsample --overwrite -d $total_stars -l 20 -r"

I've moved the orientation analysis so it's only called for SOLVED images.

Exiting gracefully is difficult, Ctrl-C works fairly well as long as it is not solving. I'm not sure how to keep the loop operational but still accept input from the keyboard.

Also Stellarium disconnecting causes it to crash and leave temporary files. On the other hand the temp files are in /tmp so I don't worry overly about them.

Anyway, updated script is in the usual place:

http://www.ebi.ac.uk/~aje/dist/starhop.pl

Thanks,

Anton

[Black Knight]

Anton, I'm having a problem with previously solvable images running out of CPU time. Did you add a timeout to the script? This is important to me as I'm running Ubuntu in a VM, so performance is unavoidably affected. Perhaps a command-line variable I could use?

[badgers]

It's going to be one of two things I'd say.

It's set for 20 seconds "-l 20" but it's likely that it is the downsampling rate that's causing the issue, try --downsample=2.0 and --downsample=1.0 and see if it helps.

If not I'll add a parameter for the timeout so you can increase it over 20 seconds.

I'm pretty sure the timeout was 20 secs in the old script, but I increased the downsample to 3.0 instead of 2.0 due to the high-resolution of my images.

3.0 may be downsampling too much for the 314L especially if you are shooting binned images.

Anton

[themos]

http://affy.blogspot...k/p5be/ch13.htm

Example: How to Handle a Signal

You can cause Perl to ignore the Ctrl+C key sequence by placing the following line of code near the beginning of your program:

$SIG{'INT'} = 'IGNORE';

You can restore the default handler like this:

$SIG{'INT'} = 'DEFAULT';

If you need to ensure that files are closed, error messages are written, or other cleanup chores are completed, you need to create a custom INT handle function. For example:

sub INT_handler {

# close all files.

# send error message to log file.

exit(0);

}

$SIG{'INT'} = 'INT_handler';

If the Ctrl+C key sequence is pressed anytime after the hash assignment is made, the INT_handler function is called instead of the default handler.

[JamesF]

I'm not sure how to keep the loop operational but still accept input from the keyboard.

select() and friends are the usual way to achieve this.

James

[Black Knight]

It's going to be one of two things I'd say.

It's set for 20 seconds "-l 20" but it's likely that it is the downsampling rate that's causing the issue, try --downsample=2.0 and --downsample=1.0 and see if it helps.

If not I'll add a parameter for the timeout so you can increase it over 20 seconds.

I'm pretty sure the timeout was 20 secs in the old script, but I increased the downsample to 3.0 instead of 2.0 due to the high-resolution of my images.

3.0 may be downsampling too much for the 314L especially if you are shooting binned images.

Anton

Thanks Anton, I'll try this.

[Black Knight]

It's going to be one of two things I'd say.

It's set for 20 seconds "-l 20" but it's likely that it is the downsampling rate that's causing the issue, try --downsample=2.0 and --downsample=1.0 and see if it helps.

If not I'll add a parameter for the timeout so you can increase it over 20 seconds.

I'm pretty sure the timeout was 20 secs in the old script, but I increased the downsample to 3.0 instead of 2.0 due to the high-resolution of my images.

3.0 may be downsampling too much for the 314L especially if you are shooting binned images.

Anton

--downsample=2.0 sorted it out, thanks!

[badgers]

I've added signal handling and graceful exit now.

Tapping 'q' will nicely quit

A Ctrl-c or breaking the stellarium connection also gracefully dies.

Fixing the 64bit portability bug now and setting down sample default to 2.0

Repost in half an hour or so

Thanks for the SIGINT pointers

A

[badgers]

The new version is up now.

I've hopefully covered everything asked for and fixed a few bugs.

I hope I've not added any new problems:

http://www.ebi.ac.uk...dist/starhop.pl

Usage Parameters are changed slightly:

--all Solve all images in directory not just new ones (default off)
--noplot Dont make solve plot images (much faster, default is plotting on)
--dir=X Look for images in the X directory
--port=Y Port Number for Stellarium Connection (10001 default)
--o=Z Solved Images always output to this file Z (default is to create individual new images)
--stellarium Wait for and connect to Stellarium (default is off)
--downsample=W Downsample images by W fold (faster, default is 2)
--stars=Z Solve on Z brightest stars (default is 500)
--delay=P Allow P seconds of solving before calling UNSOLVED

I'll talk to the mods about editing the first post here with updated details for the latest versions and changes.

One final issue is that I think the system will die on it's butt if you specify images with spaces or strange characters within the filename.

I'm too knackered to try and fix this solidly now

[Black Knight]

The new version is up now.

I've hopefully covered everything asked for and fixed a few bugs.

I hope I've not added any new problems:

http://www.ebi.ac.uk...dist/starhop.pl

One final issue is that I think the system will die on it's butt if you specify images with spaces or strange characters within the filename.

I'm too knackered to try and fix this solidly now

I came across the filename with spaces thing myself yesterday. The script just hung . I just renamed my files . I'll try this out later on.

[JamesF]

I hate it when people put spaces in filenames. It makes everything so much more complicated to process.

James

[Black Knight]

I hate it when people put spaces in filenames. It makes everything so much more complicated to process.

James

. I've tried the script now Anton, it seems to be working fine

[Black Knight]

Quick question:

Earlier I did a test solve of M101. On the jpg the script produces for the solve, RA and DEC are described as RA:210.9 and DEC: 54.37. However, my understanding is that M101 is at RA 14:03:37 DEC +54:17:29. I is all confuzzled

What am I missing here?

[themos]

RA is given in degrees (0-360) or hours (0-24).

210 degrees is 210/15 = 14 hours

.9 degree is 0.9/15 hours or 3.6 minutes or 3 minutes and 37 seconds.

The declination is given in degrees but it can be decimal degrees or degrees:minutes:seconds

.37 of a degree is .37*60=22.2 minutes or 22 arcminutes + (0.2*60=12) arcseconds, so 54.37= 54:22:12