Calibration and align batch operations

[pfile]

another buglet - i collected more frames on a target; opened up my project, started up the batchpreprocessing script, deleted the old lights and added in the new ones, started the script.

script apparently deleted everything in the "registered" directory before writing the new files; only the new files are in that directory and all the old ones are gone.

 the "calibrated/light" and "calibrated/light/debayered" directories correctly contain all the files, old and new. so all is not lost, i can run StarAlignment to regenerate the deleted files.

[Eddy Timmermans]

Hi all,

Troypiggo suggested I post my question about Bpp here.
I had some problems using this tool but after trying a couple of times, I managed to find the error I made.
Instead of using prestacked master bias and flats (from DSS), I let the tool make these master files from scratch.
But now I have another problem. While trying to figure out what went wrong the first times, I used stacks of about 10 images for bias, darks, flats and lights. And that worked fine.
Now, when I try to stack more than about 25 lights, the process ends up not finding enough matching star pairs. And after 25 retries I had to manually abort the process.
Is there a way of using some sort of parameter to end this, after say 10 tries and move on to the next image? I used several different reference images so I don't think that will be the problem.
Or is it possible to make stacks of 20 lights and stack these stacks, if you understand what I'm trying to say ?
I really want to use PI for the entire process, not only post-processing. And using the information on this forum I'm quite confident, that will be possible.

Eddy

[Eddy Timmermans]

This is a screen shot of the problem.
Apparently it has to do with the fact that the image was not guided and wide field.
But since DSS can handle the stack, PI should also be able to. So can I tweak some parameters in order to stack them anyway ?
I tried Linear Fit instead of Winsorized Sigma an changed the sliders a bit, and that seemed to work.
But after a while the problem came back.

Eddy

[kwiechen]

Hi,

There is a problem with this script: the flat frames are classified by exposure time. IMHO, there isn't any reason for doing this. And it's a big problem if you have skyflats of varying exposures, as the script won't make any master flat frame because each single flat frame has a different exposure time.

Could this be corrected, please?

Best regards,
Vicent.

This is a small problem in the GUI code - the flat frames are internally grouped by FILTER only.

 line 2182 in BatchPreprocessing-GUI.js should be

      if ( frameGroup.exposureTime > 0 && frameGroup.imageType != ImageType.FLAT )

to fix this.

Kai

[kwiechen]

another buglet - i collected more frames on a target; opened up my project, started up the batchpreprocessing script, deleted the old lights and added in the new ones, started the script.

script apparently deleted everything in the "registered" directory before writing the new files; only the new files are in that directory and all the old ones are gone.

 the "calibrated/light" and "calibrated/light/debayered" directories correctly contain all the files, old and new. so all is not lost, i can run StarAlignment to regenerate the deleted files.

The problem may be in line 960 in BatchPreprocessing-engine.js

 SA.overwriteExistingFiles = true;

This var should be set to 'false' (which is the default of StarAlignment, and the original mono CCD script uses the default)

Kai

[kwiechen]

This is a screen shot of the problem.
Apparently it has to do with the fact that the image was not guided and wide field.
But since DSS can handle the stack, PI should also be able to. So can I tweak some parameters in order to stack them anyway ?
I tried Linear Fit instead of Winsorized Sigma an changed the sliders a bit, and that seemed to work.
But after a while the problem came back.

Eddy

Can you please upload these files?

Kai

[pfile]

another buglet - i collected more frames on a target; opened up my project, started up the batchpreprocessing script, deleted the old lights and added in the new ones, started the script.

script apparently deleted everything in the "registered" directory before writing the new files; only the new files are in that directory and all the old ones are gone.

 the "calibrated/light" and "calibrated/light/debayered" directories correctly contain all the files, old and new. so all is not lost, i can run StarAlignment to regenerate the deleted files.

The problem may be in line 960 in BatchPreprocessing-engine.js

 SA.overwriteExistingFiles = true;

This var should be set to 'false' (which is the default of StarAlignment, and the original mono CCD script uses the default)

Kai

thanks, i should have read the source.

[papaf]

Hi all,
I tried to use this script with some images coming from a friend's QHY8. Since this camera doesn't need any dark and he didn't do any bias or flats, I only have lights to work with. The script fails in this condition, telling me it cannot find darks.
I know the central point of this script is to calibrate lights, so running it without calibration files may seem useless. However, what I really like about this script is the DSS likeliness, automating the whole process of alignment/stacking.
My question is: is it possible, someway, to use this script without any calibration files?

I hope this is the right place to ask for this. If not, please point me in the right direction.

Thanks!

[jeffweiss9]

   Can someone mention what the script is asking for when the BatchPreprocessing script v1.22 fails ("couldn't execute in global context...") because it says that it's "searching for a Master Dark of 2sec exposure" presumably to match the exposure time of my 2-sec lightbox flats.  Most of the time, it works to completion without this message and this only seems to happen when I have binned 2 RGB frames.  Are we supposed to include short-exposure darks that go with the exposure time of the flats in addition to the long-exposure darks that are matched to the exposure time of the lights?  When it works, it does so successfully using only the long-exposure darks to calibrate the flats.
   As a work-around when this happens, I create the master flats the old way with ImageCalibration and ImageIntegration (provided I supply a few missing FITS headers that the script requires to use the master flats), but I would love to  understand what is happening in the first place and be able to use the script to completion all of the time.
   Thanks for the help.
-Jeff

[kwiechen]

Hi Jeff,

can you please give me a console log or screenshot?


Kai

[Shelyak]

Hey Astronomers

I'm  completely new in this hobby  so I've some questions.
The first: I've downloaded the BatchPreProcessing Script from page 29 and my PI says: A Version newer than V.1.7.6. is required. But if i check my Pi version the program says that it is on the newiest version.

And then the 2nd question is how do i install the scripts or where I've to copy them?

I followed this tutorial: http://pixinsight.com/tutorials/master-frames/en.html to step 1: By creating a Masterdark PI gives out 3 Images called: Gray_Integration, Gray_rejection low and Gray_rejection high. Which oh these images is my Masterdark? And analogue which of these images will be my master-bias?

Kind regards from germany

Martin

[kwiechen]

To use the bpp script you have to install all updates.

The master dark frame is 'Gray_Integration'

Kai