IC5070

[Jules]

Hi / Hola

My favourite place within the the Pelican Nebula, I have been experimenting with the ATrous wavelet Transform rather than using ACDNR for noise control.

An hours worth of Ha data. I just love the spinderly dust fingers on this image. If/when we get some decent nights I will endeavour to get some more Ha and OIII.

Regards

Julian

[Nocturnal]

Nice detail and great star shapes too! I did notice quite a few hot pixels though.

[Jules]

Nice detail and great star shapes too! I did notice quite a few hot pixels though.

Hi Sander

Thanks for taking the time to have a look at the picture.

I am still a beginner at this sort of thing, I am struggling to see a hot pixel! I calibrate in MaximDL using a master dark taken a week or so before this image. The master dark is made up of 16 subs.

Would this be a relic from image integration?

Regards

Julian

[Nocturnal]

Hi,

they should be easy to see in your original image but even in the jpg compressed file you posted I see a lot of tiny bright specs. They aren't stars. It seems your master dark does not match your lights. I recommend fixing hot pixels before stacking if your alignment is good or when you have only a few subs. PI has a hot pixel tool but I haven't used it so someone else will have to assist you with it.

If you have lots of subs and use dithering then you can stack with a sigma clip type algorithm to exclude the outliers (hot pixels in this case).

I'm not sure what you mean with your last question but image integration did not -introduce- these artifacts. They were in your lights and darks and the calibration steps did not eliminate them. The stacking then didn't get rid of them either.

[Jules]

Sander

I will go back to the originals to see what is going on, thanks!

Regards

Julian

[Nocturnal]

Don't forget to look at the calibrated subs before they get stacked. Take a single hot pixel and compare values between light, dark and calibrated. Leave out flats for now, that'll just complicate matters.

[Jules]

Will do, many thanks!

Regards

Julian

[avastro]

Hi Jules
It's quite easy to remove hot pixels. MaxIm propose a spécific tool to do that, menu ->  filter_> Kernel filters -> Hot Pixels.
To do that quickly in many sub use menu -> view _> batch process window.
To avoid hot pixels you can do dithering during your acquisition, and or combine using SD mask.
Cheers

Antoine
http://astrosurf.com/avastro/

[Jules]

Hi Jules
It's quite easy to remove hot pixels. MaxIm propose a spécific tool to do that, menu ->  filter_> Kernel filters -> Hot Pixels.
To do that quickly in many sub use menu -> view _> batch process window.
To avoid hot pixels you can do dithering during your acquisition, and or combine using SD mask.
Cheers

Antoine
http://astrosurf.com/avastro/

Thanks Antoine

I am just creating some more darks, I will have a look in Maxim with your suggestion.

Regards

Julian

[Juan Conejero]

MaxIm propose a spécific tool to do that, menu ->  filter_> Kernel filters -> Hot Pixels.

A 'kernel filter to fix hot pixels' is a simple median filter with a 3x3 structure (either a square or a cross structure), maybe applied through a mask to protect the rest of the image. This is easy and accurate with MorphologicalTransformation in PixInsight.

Better than that is applying a specific tool to fix hot pixels in PixInsight:

- CosmeticCorrection script by Nikolay Volkov.
- DefectMap tool by Carlos Milovic.

To apply one or more processes to a set of images, ImageContainer can be used in PixInsight, when necessary.

In general, hot and cold pixels should be fixed through accurate image calibration. Cosmic rays and other artifacts are naturally removed during image integration with pixel rejection (sigma clipping, etc.), including any surviving hot pixels, provided you apply dithering between exposures.

[Nocturnal]

Image calibration will not remove hot pixels for cameras that use cooled Sony CCDs such as the QHY8 and M25C. These are calibrated without darks. In my case I generally dither to allow stacking to get rid of them.

[Juan Conejero]

That's true. However, wouldn't it be a good idea taking some dark frames with these cameras, just to use them as defect map images?

[Nocturnal]

Indeed I have a dark frame library that I use to find and fix hot pixels. But that doesn't mean the dark frames are used during calibration in the traditional sense, subtracting them from lights. I use a master bias for that.

[Jules]

I have a QSI camera with a KAF-8300 CCD chip. I had never thought of keeping my old darks as defect maps. I normally keep my master darks and bin the dark subs.

Modified for my bad English!!

Regards

Julian

[Nocturnal]

well once you have a master dark there is no reason to keep your individual darks, at least not for defect maps.