Rejection - High and Low

[stevek]

Hi,
Please see attached. I can't understand why these calibrated channels have so much of a blizzard of white on the high channel rejection?  It is from a Atik 460 CCD operating at -20C.
Any thoughts please?
Thanks, Steve

[pfile]

just increase the 'high' slider a bit and iterate until you think what's being rejected is definitely hot pixels/airplanes/satellites.

every image is different; every image's statistics are different; every image will need different high and low sigma values for rejection...

rob

[stevek]

Many thanks. I did not understand that point.
Do you guys use the pre processing script or do you do calibration manually?

[pfile]

i have been using PI since before the BPP tool and so i persist in doing it all by hand.

someday i will start using the script.

rob

[Phil Leigh]

The integration done by the script is only a ROUGH guide so you have something to look at and sanity check - it is categorically not intended to be accurate. You need to manually integrate and to tweak the low/high settings for EVERY image as Rob says. Fortunately this process runs pretty quickly since all the data will be in cache.

[papaf]

Is it really that important to tweak the low settings? I find the high to be pretty straightforward, while the low either produce a black or full white calibration image...

[pfile]

well you may have some remaining cold pixels that you want to reject from the low side but i agree that the high rejection is much more important given the nature of the 'unwanted' things we find in images…

sometimes you might have a sub with a cloud or perhaps part of a tree or something that you want to reject. but generally i just reject the whole sub unless the subject is far, far away from the shadow… the uneven illumination of the frame can play tricks on the weighting logic and such a frame could be over- or under- represented in the final stack which would be bad.

rob

[jkmorse]

On the question of using the Batch tool for integration, Juan's new detailed ImageIntegration documentation clearly states that you should never use Batch preprocessing for your lights.

And when Juan says it, its coming from the mouth of the Master himself.

Jim

[Phil Leigh]

He doesn't say you should never use it... he says you will always need to tweak the Integration settings for optimum results. The light integration from the script is just a starting point...

[jkmorse]

Here is Juan's text:

"Don't use the BatchPreprocessing script to integrate your light frames. In most cases, BatchPreprocessing does a fine job for generation of master calibration frames, image calibration and registration. However, integration of light frames is a critical process requiring manual intervention to fine tune pixel rejection and image combination parameters. The integrated output of BatchPreprocessing can be used as a quick preview of the image that can be achieved, but it is not the optimal image by any means, and many times you're quite likely to get a grossly wrong result (e.g., invalid rejection of plane and satellite trails, etc.)."

Seems pretty definitive to me, IMHO,

Jim

[papaf]

Still, the only parameter I'm finding myself changing while integrating lights is High clipping, and it doesn't affect the image very much anyway.
Are there other parameters I'd better check?

[Phil Leigh]

What Juan said and what I said are exactly the same meaning. Use it but only as a guide...

[stevek]

The question is, how much better does it make your image doing manual calibration as opposed to letting the script do it all for you and fire out your completed masters?  Probably very, very little!

[Phil Leigh]

It's not the calibration - or registration - the script does those fine. Just the final integration step will probably benefit from tweaking the rejection parameters.

Speaking personally, I find that the integration step script with its default parameters does a perfectly fine job 80% of the time...