PixInsight Forum (historical)

PixInsight => General => Topic started by: FlyingBeagles on 2018 July 01 01:47:53

Title: Post Deconvolution - Still having some odd artifacts
Post by: FlyingBeagles on 2018 July 01 01:47:53

Hi All,

I am a bit of a newbie when it comes to Pixinsight (and AP for that matter) and no doubt like most who started with this amazing tool, there's a fair learning curve that goes with it. 

I recently ran a set of narrowband subs on NCG3372 using 20 each of Na, Oiii and Sii.

I've followed a few pixinsight online tut's (both manually and via bachpreprocessing to double check) to arrive at a set of master flats, bias, and darks for each filter as well as master lights for each.

I seem to be having trouble with my O and S compilations (to be honest, I wonder if I didn't mix up the filter names when taking the shots) but am pretty happy with my Ha.

The only problem is, I am getting a stack of artefacts (or what appear to me to be artefacts) in all my master lights and they remain in my deconvoluted Ha (as tiny white dots).

I've attached all three in the hope someone might be able to say, "yep, that's xxxx, you need to yyy" to fix it.

https://www.dropbox.com/s/cygo25n43k4kqml/LIGHT_HA_MASTER_DECONVOLUTED.xisf?dl=0 - This is the deconvoluted Ha

https://www.dropbox.com/s/axx96jc8pk0bacd/LIGHT_OIII_MASTER.xisf?dl=0  - Master Oiii Light

https://www.dropbox.com/s/04ir9u6z6shbqgb/LIGHT_SII_MASTER.xisf?dl=0 - Master Siii Light.

Any help would be thankfully received and gratefully applied.

Pete

Title: Re: Post Deconvolution - Still having some odd artifacts
Post by: pfile on 2018 July 01 09:14:59

looks like you need pixel rejection turned on during integration. also just a language tip - it's "deconvolved" not "deconvoluted".

rob

Title: Re: Post Deconvolution - Still having some odd artifacts
Post by: FlyingBeagles on 2018 July 01 14:32:18

Thanks mate, appreciate both the insight and the language lesson.

Pete

Title: Re: Post Deconvolution - Still having some odd artifacts
Post by: RickS on 2018 July 02 15:06:27

Hi Pete,

As Rob suggested, your artifacts look like hot pixels and turning on rejection and tuning the rejection parameters will clean them up.  If you're not already doing rejection then you should be!  You could also consider using CosmeticCorrection to clean up the hot pixels during calibration of the individual subs.  I like to attack them as early as possible and let rejection take care of any remaining stuff. If you're not dithering your subs then that's another improvement you can make.

Cheers,
Rick.

Title: Re: Post Deconvolution - Still having some odd artifacts
Post by: FlyingBeagles on 2018 July 03 02:29:48

Thanks, Rick

Yep, I am using pixel rejection but not sufficiently it seems (likely a product of still learing what works etc).  I will try and CosmeticCorrection as well and see how it goes.

Pete

Title: Re: Post Deconvolution - Still having some odd artifacts
Post by: pfile on 2018 July 03 08:21:11

lower the sigma value in the high rejection setting until they go away... somewhere between 3-4 usually works for me.

rob

Title: Re: Post Deconvolution - Still having some odd artifacts
Post by: FlyingBeagles on 2018 July 03 13:54:07

Thanks Rob,

WIll give it a go.

Pete