Author Topic: Master Flat Creation  (Read 1931 times)

Offline Jack Harvey

  • PTeam Member
  • PixInsight Padawan
  • ****
  • Posts: 975
    • View Profile
    • PegasusAstronomy.com & Starshadows.com
Master Flat Creation
« on: 2010 September 20 11:07:18 »
This process, although very exacting, is tedious enough to discourage the creation of Master Flats in PixInsight.  Master flats must be created for each image so this is a process used frequently, unlike creation of Master Bias and Darks for a library.  The use of two separate tools to perform this process is extremely cumbersome.  I like the Flat Calibration routine and if that would flow into the production and saving of the integrated and labeled Master Flat all would be well.  But to go back and find the calibrated flats and run through a integration and then label and save the integrated master in the appropriate folder is too much.  This process, should with a single click, deposit a integrated and named master flat into the same file from which the raw flats were obtained IMO.

I am working on a couple of images today and by the time I have created the Master Flats for 5 filters for two images I am to burnt to continue processing the raw image data:-(

Having said all that, I have done a side by side visual comparison of the calibrated light frames and the ones calibrated via PI are certainly preferrable to my eye over the same frames calibrated using another program. Lets just make it more user friendly<G>.
« Last Edit: 2010 September 20 11:27:12 by Jack Harvey »
Jack Harvey, PTeam Member
Team Leader, SSRO/PROMPT Imaging Team, CTIO

Offline William McLaughlin

  • Member
  • *
  • Posts: 58
    • View Profile
    • Images from the Night Sky
Re: Master Flat Creation
« Reply #1 on: 2010 November 02 17:03:30 »
OK, this reply is less about making a master dark than using one in the calibration routine but this seems like a good place to post. The following makes no sense to me:



The above would seem to indicate that one can bias and dark subtract the master flat before use. Bias I get, those are pretty much unchanging for a given camera so the one used elsewhere would be fine. Dark subtract makes little sense to me. Is it using the master dark frame specified for use with the light frame (object) data and scaling it? The length of the dark for the flat is almost never anywhere near that for the light (so one can't use that w/o scaling, thereby possibly introducing unnecessary error) and there is no place to specify a master dark frame for the flat or dark frame set (to generate a master dark) for the master flat.

In any case, better flats are generated by using flat specific darks and multiple component flats (which are easy, quick, and normally automated so why not do them?). Therefore the master flat would probably be created ahead of time and already dark and bias subtracted (in which case one would uncheck "calibrate" rendering this academic).

It seems to me that if one is going to offer the option to calibrate, one should also provide a space to input flat specific calibration frame(s) if one has them.

Of course, even more ideal would be a setup that allowed one to specify the all the component lights, darks and bias' in each section (master dark and master flat) so that the calibration process would calibrate and generate all the calibration frames from the multiple component files and then proceed to perform the calibration on the light frames as well. MaxIm does something like this (but with less control than I like). I assume one could use the image container and a set of processes to do this, but since this is such a common robot-like task and normally these items are all linked, it would seem to me a good idea to have this all in one step as a process.

So what it looks like to me is that, as is, one might as well separately generate and calibrate one's calibration frames (master darks and master flats) and just "uncheck". Is that correct?
« Last Edit: 2010 November 02 17:32:35 by CCD-PIX »
Website: http://nightskypictures.com/

Observatory: http://nightskypictures.com/raptor-ridge.html

For every complex problem there is an answer
that is clear, simple, and wrong.

H. L. Mencken