Calibration and align batch operations

[kwiechen]

0.0.7f adds some improvements to the user interface.

[RobF2]

I've been unable to give this script a run on XP32 unfortunately.  As Troy has commented earlier, you can't create suitable Master frames without using the script itself it appears, however when I try to combine a reasonable number of frames for bias or darks I run out of memory. 

The usual way around this is to reduce the default Buffer Size during image integration from 16mb downwards.  Would it be possible to expose that parameter to the script.

I'm quite disappointed - PI REALLY need these sorts of facilities currently, so thanks for all the hard work!

[mmirot]

Kai,

The light frames file don't show up after selecting them.  Bias, darks, flats select fine.

Max

[mmirot]

Never mind. I think the problem is on my side not your code. 

Max

[kwiechen]

Rob, Max,

thank you for testing this.

@Rob: the script was tested with Win7 64bit. I will test with winxp 32 and of course we can expose more parameters to the script.

@Max: can you give me some hints or data about the rotator keyword support. If possible I can merge this into the current version.

Kai

[kwiechen]

Hi Rob,

I have tested with 30 bias raw frames binning 1 from my QSI583ws without any problems using Windows XP 32bit/2GB RAM.

Kai

[mmirot]

Rob, Max,


@Max: can you give me some hints or data about the rotator keyword support. If possible I can merge this into the current version.

Kai

Sure,

I don't want go forward with rotator code just yet. I found that CCDAP5 with MaxIM DL appears have round off error occasionally after re- initializing the rotator.
That is the one image set may be a degree off under the Rotator key word,  i.e now you have 354 and 355 degrees on frames that are theoretically identical rotation.
This won't work well !


MaxIM DL inserts a FITs key  Pierside ' East '   or ' West '
Also the there is Flipstat key word.
These might be better since East verses West or Flipped verses not  are all we need.

Another approach is to have the function work through the image title.  The non case sensitive word East or West in the title  ... M51_East ... M51east etc.

[mmirot]

1
 I found a problem with the imagetyp Key word.  My recent set of images taken with MaxIM DL did nt get recognized because the contained imagetyp ' light'  rather than ' light frame '

I am sure that this is due to a update/ with setting changes in my software. Surely, this will cause problems for someone latter on too.  Can we make it less restrictive so both work?

2

7f the clear function with light frames did not work.

I hit reset the files go away but then you can't open a directory and add them in again.

Max

[kwiechen]

Max,

flexible handling of IMAGETYP values is on my TODO list (for example to use IRAF conventions).

The clear function is unfortunately broken. I will try to fix this.

It is possible to group files following different sets of FITS keywords, therefore there should be no problem to use 'Pierside'

Kai

[mmirot]

Yep had the IRAF option checked. No longer a problem.

Max

[RobF2]

Hi Rob,

I have tested with 30 bias raw frames binning 1 from my QSI583ws without any problems using Windows XP 32bit/2GB RAM.

Kai

Hi Kai

Yes, I was being greedy - using about 41 frames bias and darks.
One other problem - the script seems to crash if you don't specify any flats or master flat.  Surely should be an option that don't have to specify flats?

(BTW, I was able to get the script going happily using my dual-boot Ubuntu 64bit setup)

[mmirot]

Rob,

You will be a lot happer with a 64 bit system with at least 6-8 GIGs of memory. I makes a big difference. 32 bit is limits ram access to 2 GIGs

Max

[RobF2]

Yes, definitely on the wish-list to upgrade Max, but feeding the family comes first 

[astrospotter]

After looking at this a month ago it was 'nice' but still so very easy to make an operator error or in several cases to figure out in an intuitive way.

Now with version 7f my hat is off and I thank you SO much for the efforts of putting together this extremely valuable and key piece of missing workflow to PixInsight.   Looking much better now.   I see you have the keyword recognition to avoid operator error and the layout as well as logic in the tool checkboxes is far more intuitive.

I had a problem with what looks like hot pixel issues so will have to revisit in case I operated it wrong but the workflow and so on is much improved.  I'll try to sort out what went wrong.  I went back to MaximDL and figured out where I had gone wrong there (had skipped an align that I thought was automatic but required that step separately).

Over the last few days I have been struggling with both MaximDL and DeepSkyStacker both of which have had assorted serious issues that have led to inability to form masters of quality.

IN ANY CASE, THANK YOU!!!     This will make a truly important contribution to everyone's workflow.

Mark Johnston

[pfile]

am i correct in assuming that this script will not work for a DSLR user since it relies on some set of FITS keywords? thanks.