Author Topic: calibration and preprocess modules  (Read 4904 times)

Offline mmirot

  • PixInsight Padawan
  • ****
  • Posts: 881
calibration and preprocess modules
« on: 2009 April 24 11:13:10 »
Juan has suggested this may be in the master plan.

I think the Pixinsight would do better than most products out there and would really be added attraction to future owners of Pixinsight.

I think full work flow after acquistion should be  considered.

Thoughts

1. Module to review and sort subframes.  This could be done by manual review, automated or semiautomated methods.

Basically, we all need to sort the good from bad.
Measurements such as FWHM and SNR would be helpful.

2.  Calibration module with a data base.

 It would be nice to match up flat, dark, bias by the fits header paramenters.
The best approach is making a data base so we could select parameters
 instrument,  , bin,  duration, date , angle rotation( flats), date,
 etc

For example, here is common acquistion sequence at my observatory

Finish an acquistion run over serveral nights , a week or a month apart.
It contains perhaps three objects.
Now lets make it complicated.
The scope has a field rotator.
One object is OIII, HA, SII narrow band, the others are LRGB
Subframes were acquired over many nights. Likewise the flats had be acquired at different times since I have so many filters and my shutter is slow.

Remember for my flats alone I have to acquired flats 3 objects, one using 3 filters and the other 4. I also have six position angles. That is 24 flat sets just to match up with various light frames/filters over multples nights!
Let us say it took me 5 nights to get the full image sets for each side of the meridan with the proper angle on these objects. I have a few flats from last month on two filters with same angle on some, the rest I got this month but it took 6 night to complete.

Now I need a way match up this mess!
My  automation software does nice job of acquiring this image plan.
The fits headers contain the needed  info including a plate solution with a position angle.

Now , I run the data base for one object " M100".   The filter, temp, bin, dates, angles are ID from the hard drive. I can run the data base to select appropriate flats, darks, bias.  

So far no software has done a really good job of making this easy.  MaxIM DL has sort of done this half way but does not have a way to scan a HD and easy select images with various parameter such these accoss dates etc.

Max Mirot

Offline Nocturnal

  • PixInsight Jedi Council Member
  • *******
  • Posts: 2727
    • http://www.carpephoton.com
calibration and preprocess modules
« Reply #1 on: 2009 April 24 14:23:59 »
I have to admit that I'd rather have PI focus on post processing but I'm biased as I use DeepSkyStacker for stacking and it does an excellent job.

I think one of the problems with Maxim is that it's a one-stop-shop. Too big, bulky and expensive. I would not want PI to morph into that. It's OK to leave some areas to others and go deep on the remaining.

Luckily PI is fully modular so someone can write PCL routines for image calibration and organization and sell them separately :)

Of course I realize that one of the other powers of PI is multi platform support. I'm sure Mac and Linux users would be delighted to see a true stacking solution in PI.
Best,

    Sander
---
Edge HD 1100
QHY-8 for imaging, IMG0H mono for guiding, video cameras for occulations
ASI224, QHY5L-IIc
HyperStar3
WO-M110ED+FR-III/TRF-2008
Takahashi EM-400
PIxInsight, DeepSkyStacker, PHD, Nebulosity

Offline mmirot

  • PixInsight Padawan
  • ****
  • Posts: 881
calibration and preprocess modules
« Reply #2 on: 2009 April 25 12:37:23 »
It is not my number one wish.

Someone mentioned it as future possibly beyond ver 1.5.
Juan said if and when they go there they hope to offer something better.  

However, I think calibrate a selction features would be a nice addition and would attract additional users.
It  is also a logical extention now that we have registration and intergration.

The data base feature I mentioned would  probably not that attractive to color camera users.
The file complexity is greatly reduced. That is one of the reason just bought a QHY8 to play with. :)

I hope this simulates discussion for future planning.
I own a lot software so comparisons are natural

Max