Dark frame artifact removal

[kinetic]

Hi all,

I had a go at Eta Car last night before the moon came up.
Not a serious attempt, but the seeing seemed to be fairly good.

I get these dark frame artifacts that seem to survive the
dark frame subtraction process, and only show up when doing
extreme histogram adjustments.
That's probably to be expected though ( my dark frames can never
be an exact exposure equivalent of the lights, at least in my DSI
because it is uncooled).
But having said that, most are gone after the subtraction, but
this bugger seems to get through.
When I looked at my dark frames, the culprit is a pixel that turns
on and off at random.

Is there a technique that deals with these pixels?....

Or is it easier to deal with it via a post processing 'touch up'?

Does anyone have this happen?

thanks in advance,

Steve

[gregbradley]

The healing tool in Photoshop will handle that.

Greg.

[iceman]

Yeh just clone it out.

[theodog]

Nice image Steve.

I have had many problems with dark subtraction in the past, dark blobs and light pixels.
Unfortunately the only way to truely do away with them was to take as many darks as I could, the more the better, and median them together to make a master dark. If you can take 2-3 lights and then 1-2 darks. Perhaps save the temp and exp length in the master filename, it could be used for future similar images and save time. (Cover the scope on moonlit/cloudy nights and build up a library of them), something to get a camera fix during these times.:P

Also use bias frames -0sec length exp's and treat them the same way.
Post 'healing tools' can alter data.

Unfortunately there is no real quick fix.
:D

[kinetic]

Thanks Jeff, Mike and Greg,

I certainly try to take about a ratio of 1/3rd darks to
lights, always.

Some nights it's almost non-existent, some nights bleeding
obvious.

I did read somewhere that just putting the big (plastic) dust cover
over the end of the OTA wasn't good enough because IR light
goes through plastic to a certain degree, and alfoil over the
CCD camera was best.
Not sure how much leakage would contribute that way, but maybe it's
worth a try.

Also, my dome isn't exactly pitch black darkness either....I have LEDs,
power supply neons...monitor LEDs...all taped over with black tape,
but there is probably a measurable leakage through the BACK end of the
OTA of this feeble light.

What I shouldn't be offering as my example is last night's shot...after all
the moon was coming up in half an hour more, the dome was still cooling,
the OTA was still cooling...too many variables, hardly a scientific approach.

thanks for your help guys,

Steve B.

[rat156]

Is there a hot pixel removal tool you can use? If you run that through the subs before combining them that should remove the offending pixel.

Cheers
Stuart

[Geoff45]

Nebulosity (http://www.stark-labs.com/nebulosity.html) caters for this. Good software and really cheap.
Geoff

[kinetic]

All sorted guys, thanks for the help.
I had the 'threshold' setting for my dark frames set wrong.
It was dealing with all but the brightest ones.

No sign of them at all now.

Steve