Unofficial documentation for ImageCalibration

[bitli]

Tired of waiting for documentation I attempted to write one for ImageCalibration. It is attached here in two formats: The original pidoc and the generated html. You can regenerate the documentation from the pidoc sources or simply unzip the html in the directory C:\Program Files\PixInsight\doc\tools or equivalent (this may require administrator privilege).

BEWARE: I copied and adapted some text from other documents and from the tooltips (the copyright is assigned to Pleiades Astrophoto, so I hope they do not mind). I also wrote some parts myself and inserted information I found in the source file.  I hope that it is correct, but it may contains errors.  Also I am at an astronomical distance of having the talent of Juan for documentation, however I hope that it is better than nothing.

You can post suggestions or corrections in this thread, I will update the attached files as time permits.  Is somebody is willing to correct the English in the pidoc source, I would be very grateful.

-- bitli

[georg.viehoever]

I love that you are trying to do this .

At the same time, against all experience, I am still hoping to get decent documentation from the development team...

Georg

[Philippe B.]

Hi Jean-Marc


Very nice job  8)


You are now ready to make all Pixinsight documentation !!!! 
Juan could hire you for this 

[Juan Conejero]

Jean-Marc,

Great job. Let me "digest" this before answering. You're taking a serious risk with this initiative... 

[bitli]

From georg:

I am still hoping to get decent documentation from the development team.

The problem is they do not do decent documentation, only outstanding documentation - or none at all.  I just think that half decent may be good enough some times... We'll see

From Philippe:

You are now ready to make all Pixinsight documentation !!!!

well, I retire in a few years, maybe if the doc is not finished by that time 

From Juan

You're taking a serious risk with this initiative...

You are taking serious risk in trying to digest it 

-- bitli

[jerryyyyy]

Thanks, this is great stuff to read on a plane.  I wonder if you can help lazy novices like me with one question.  In the BatchPreprocessing Script if you start with raw bias darks and flats etc, BPP will create masters.  I sorta discovered this by chance and it seems to cut down a lot of effort.... but am I missing something and getting less than optimal results? 

[georg.viehoever]

...
From Philippe:

You are now ready to make all Pixinsight documentation !!!!

well, I retire in a few years, maybe if the doc is not finished by that time 
...

You probably just created a retirement employment opportunity for yourself 

[Geoff]

Very nice Jean-Marc.  Only another 61 to go.
Geoff

[mads0100]

Jerome,

    I found that using the BPP is giving me poorer results than doing it manually.  Specifically with my flats.

[bitli]

BPP tends to "calibrate" the flat with the image darks (at least it did a couple of weeks ago).

You may want to look at the process of BPP (using the log) and try to apply exactly the same manually.  It should give exactly the same result (pixel to pixel) at least as far as calibration is concerned (SA has some statistic in the algorithm).  Try on a small set of images, this is just to understand what happens.

If it doesn't give the same result, then you did not yet found what BPP is doing.

There is no magic in BPP and no image related intelligence, it is "just" a pretty clever way to sequence a set of standard PI operations.  Clever but sometime surprising.

-- bitli

[mads0100]

I found that I couldn't use the same settings in BPP as I could manually, specifically for integration.  For instance, you can't set 'equalize fluxes' for your flats.  Unless that's behind the scenes?

[jkmorse]

Bitli,

Great stuff!!  I can't wait to dig into your suggestions and play with this since ImageCalibration is one of the most important but least documented parts of PI. 

Thanks for a great effort.

Jim

[bitli]

mads0100, your are right, the BPP script does not allow you to set all options that you can set manually.  The other way is always possible (you can do manually what was done with BPP).
Unfortunately the only (but instructive) way to find what BPP is doing is to read the sources, although the logs provide good information too.
-- bitli