Issues with BatchPreProcessing

[Eddy Timmermans]

Yesterday I tried to stack 241 lights with 60 darks, 40 flats and 100 bias frames.
I checked the images with the SubFrameSelector tool and found they had FWHM values between 4.6 and 3.3.
But when the script tried to stack everything, I got an error that no stars were detected.
The reference image has a FWHM value of 3.3.
I noticed that during the script debayering an calibrating, the noise levels were all zero.
I then tried with a reduced stack of 10 images per frame, but that didn't work either.

So for the moment I'm back to DSS for the stacking.

Since I managed a similar stack a few weeks ago, something must be wrong.

So what can I check in order to find my error ? Where do I begin ?
The only difference is that this time I used a 200mm lens instead of my refractor.

Thanks for any suggestion,
Eddy

[Eddy Timmermans]

I would also like to know if it is possible for the script to write to a log of some sort.

That way I could be more specific on where things go wrong.

Eddy

[jkmorse]

Eddy,

Hope you get to a solution.  I had the same issue recently with some NB stacks that I am still trying to resolve (see this post of the problem: http://pixinsight.com/forum/index.php?topic=6072.msg41227#msg41227). 

My images are monochrome CCD images rather than bebayered, but we seem to be having the same issue.  One idea I have had is that it may have to do with the dark frame scaling.  I have been using a standard 30 minute dark master and while that created no problems in the past, is the only thing I can think of that I haven't tried.  As such, I am in the process of shooting new darks that match the time and temp of my lights exactly and I will see if that helps resolve the issue.

I will post here once I get a chance to test.

Jim

[Cosmick]

I would also like to know if it is possible for the script to write to a log of some sort.

Eddy

Every operation performed during BatchPreprocessing is written to the Process Console. You may need to increase the limit on the number of lines the console displays when working on a large stack of images. You can do this in the Global Preferences under the Miscellaneous Process Settings. Set to zero for unlimited lines(mouse-over for info box).

[Eddy Timmermans]

I noticed that my flats are OK individually, when I stretch one usign STF I see a nice round glow and darker edges. But when I stretch the FLAT_BINNING file, I get a black image. I will check again this evening.

Perhaps this is the source of the problem.

Eddy

[jkmorse]

Eddy,

I had the same thought about the flats and one of the reasons I am shooting new darks.  It may just be too much to ask the system to scale down a 30 minute dark master to fit a 10 second flat stack.  But just in case, I am also shooting matching darks for my lights too.

Good luck,

Jim

[Eddy Timmermans]

My darks match my lights (60 secs). My flats were 0.3 secs.
But last time this wasn't a problem.
Last time when I looked at de FLAT_BINNING file, it looked OK, a round glow in the middle and darker edges, with the occasional dust particles.
I will try to make the master flat seperately using the integration tools.
Eddy

[Phil Leigh]

The scaling of the darks will not be a problem. More likely you have not checked the CFA option in the script and yet are feeding it undebayered RGB fits or CR2 files (or vice versa).

To be clear, unless the files you are feeding the script are monochrome (from a CCD or physically debayered DSLR) you need to check the CFA box.

Also if you are converting your CR2 (RAW) files to fits using the Batch Format Conversion Script, you must use the correct input format hint options: see attachment

[Eddy Timmermans]

I did check the CFA Option because I'm feeding it CR2files from my Canon 450D.
And when I first converted to fits I used the same settings as in your attachment.

I wil try again this evening using reduced sets of images.

I will also check for errors during processing and see where I can correct them.

Thank you for the feedback
Eddy

[Phil Leigh]

I would double check that you have not accidentally included a dark or bias in with the flats - that would make the master flat "black"...

[Eddy Timmermans]

I get this error message :

Opening files:
C:/Users/Eddy/Pictures/Astrofotografie/Astrofotografie/Resultaten/registered/IMG_2793_c_d_r.fit
Reading FITS: 32-bit floating point, 3 channel(s), 4290x2856 pixels: done
Normalizing sample values: done
Computing image statistics: —
*** Error: C:/Users/Eddy/Pictures/Astrofotografie/Astrofotografie/Resultaten/registered/IMG_2793_c_d_r.fit: Zero or insignificant signal detected: Empty image?
<* failed *>

However, the orignal image IMG_2793 had the lowest FWHM and shows many stars.
After Registration, this image still has quite a few stars but with very high contrast. No background detail visible.
After debayering, nothing is left.

I use standard settings.

Where do I go wrong ?
Eddy

[Eddy Timmermans]

Sorry I ment after calibration the reference image has very high contrast, not after registration

[Phil Leigh]

1) just put the LIGHT frames into the script... no bias, dark or flats
2) if that works try again adding in flats then bias then finally darks... see at what stage you get a problem.

Check your master flat, bias and dark as they get created at each step above to make sure they look correct.

[Eddy Timmermans]

It is not possible to just use lights. I have to include darks or flats

[Phil Leigh]

Include bias frames but no darks or flats...