Little black dots !

[Brian28]

I'm trying to process the subs I shot last week .. trying out pixinsite as well as my normal photoshop .. 

i have this Ha version of the Rosette neb and it seems to have lots of little black dots all over it ! ?..  has anyone seen this before or have any ideals .. please ..

[Brian28]

And yes , the Optics were all clean ..  ?

[david_taurus83]

What camera did you use?

[alacant]

Hi. Send a link to the file. Cheers.

[Adreneline]

I hope this is not a red-herring but I have seen exactly that effect using an osc camera when taking test shots when the camera had not cooled.

Most typically when taking flats the following morning after a night of imaging. Whilst waiting for the camera to cool I would take test exposures to determine the correct exposure time for my flats - the test exposures often had black dots all over them. Once the camera had cooled to -10 degrees the dots all disapperared.

I have also seen them when taking random shots (of distant pylons) when aligning scopes, etc. - once again without cooling the camera. My camera is an Atik414osc - I've not seen this effect when using the Atik428ex mono.

Really hope this is not a red-herring but it might be worth a thought or two.

HTH

Adrian

[ollypenrice]

Do you see these dots on the single raw subs? The first thing to clarify is the point at which they appear, so first stop is the single raw. Then try stacking without calibration, then with each calibration file tried in turn.

Olly

[Davey-T]

Zoomed right in they look like hot pixels that have been reversed in some way or else dead pixels that haven't been removed by calibration.

Do they show up as bright dots on a single uncalibrated sub ?

Dave

[geordie85]

Turn off cosmetic correction in DSS. It looks like it's set too high and its deleting stars thinking that they are hot pixels. 

[Brian28]

Hi all .. the camera is QHY163C. It was cooled to -20 , I took the flats and darks at the same time as the subs .. 

 

Georgeie85. - restacked in DSS with cosmetic correction turned off and it did the same thing ..

[david_taurus83]

They all look exactly round and are the same size. They are also heavily concentrated along the top edge. I would say its sensor related. Can you take a daytime shot and see if they are still there?

[Brian28]

I have some pics the subs were processed in DSS . Cosmetic correction turned off . 

I cannot see the spots on any image except the final processed pics. 

 

 

[Brian28]

This is a shot of the stacked light subs only no Darks , bias or flats 

[Brian28]

My camera settings were 20 gain , 80 Offset .. 

[david_taurus83]

Have you tried stacking in Pixinsight?

[Brian28]

Hi David , yes I have . But I got the same effect ..as PI is more complicated than DSS   I stacked the whole lot again in DSS 

[Davey-T]

Zoomed in to pixel level they look like stars that are only magenta in colour, are there actual stars in these positions on the image ?

Dave

 

[Brian28]

Hi Dave , no I don't think there are stars in those positions .. 

just to confuse things even more , this is an image I took last week ( you know when it was clear for a couple of night ?) Of M51 , I stacked it with subs taken on the 25 & 15th Feb ..  total time 3hrs 41mins , those taken on the 15th were taken at the same time as the pics above with the spots on ! 

 

[Davey-T]

Something similar on star edges in M51image stretched and zoomed in, don't know if it's related though.

Are you dithering or using binning of some description ?

Dave

[Brian28]

Hi Dave , I'm not getting that when I zoom in , will post pic in minute , DSS is stacking .

i am dithering and bin 1x1 

[Oddsocks]

Without access to a small set of your raw files and calibration masters to see what is going on I will hazard a guess that if the subs are from an OSC camera with an Ha filter fitted, as you mentioned in the first post, and if you were using mount dithering (deliberately or accidentally) then the dark spots are caused by the random overlay of the green and blue pixels after registration and combination, which contain nothing but noise.

It's been a while since I've processed any OSC+Ha filtered images but the way I used to approach this was to calibrate the dithered subs, extract the red pixel data only from each sub in MaxIm, then register and combine the extracted data subs as a monochrome image before upscaling and adding to the red channel of a matching RGB image in Photoshop or using as-is as a monochrome Ha final image.

I don't know DSS that well so I can't suggest a way of combining Ha filtered OSC data that ignores the green and blue pixel data and only uses the red channel to create the final image. You could try the 'Create Super Pixels from the raw Bayer matrix' option under Raw/Fits Digital Development Process Settings, this will combine the four pixel elements of each Bayer group into a single pixel of a downsized monochrome image, the disadvantage is each super pixel will contain the noise from four pixels and the lightness data from just the red, but you shouldn't have the dark spots. You would need to upscale the image in photoshop and blend it in to the red channel of a full colour image.

Alternatively, try the Bayer Drizzle Algorithm setting under Raw/Fits Digital Development Process Settings, import the image into Photoshop and extract the data channel corresponding to the red pixels and blend it into the red channel of the full colour image.

I expect there will be other users of DSS with experience of OSC with an Ha filter who can suggest the best way of handling these types of images.

As I use PixInsight I would follow a similar path to the method I used to use with MaxIm, calibrate first, extract the red pixel data only from the bayered images and go on to register and combine the red data as a monochrome image before upscaling and adding back later into the normal full colour image acquired without the Ha filter.

William.

[carastro]

I am suspicious you have accidentally got a dark mixed up with the flats and all the hot pixels are coming out in reverse.

I had an incident once when I accidentally put a sub from M51 in with flats of an entirely different image, and got a negative image of M51 in the middle of this different target.  It was the shape of M51 which was instantly recognisable that made me realise what I had done.   

This is what is making me wonder if you have a dark in as a flat. 

Carole 

[Brian28]

It appears Carole was correct , there was a problem with a set of dark frames I took .. though I don't know why ?

I used a set taken a few weeks back that had the same settings and all is fine ..  thank you all for your thoughts .. 

 

regards Brian 

[Davey-T]

Glad you're sorted Brian, nothing worse than mystery defects.

Dave

[carastro]

Glad I was able to help.  Well that's that mystery solved then.  One to chalk up for future reference. 

Carole