DMK/Photoshop colour issues

[matt]

Hi guys.

Thought I'd put a question out there in the small hope someone might have an answer.

I've been concentrating on RGB planetary imaging with the DMK 21AF04 mono firewire cam and Astronomik Type 2 filters.

I use K3CCDTools capture software to acquire my colour frames. I then run each AVI through VirtualDub to convert to BMPs, then crop and centre in ppmcentre, then into Registax for some gentle wavelet processing and save as 16-bit TIFFS, then into Photoshop 7 where I convert to Greyscale, convert to 8-bit, merge channels.

I'm noticing that my initial merged image looks to have a very strong red cast, which strikes me as odd since I stack the same number of red, green and blue frames into each channel.

What's really got me puzzled is how hard I'm finding this to correct in PS. I've tried a number of different approaches in PS...but I always end up with unnatural looking colour.

Any ideas what might be causing this? Is there something I should be paying closer attention to at capture time? I usually just work straight off a greyscale histogram in ICCapture and try to get the same for each channel. Should I be using individual colour histograms...and are there such things in IC Capture which you fire up during capture?

Am I doing something wrong in Photoshop? Maybe I need to match my Photoshop workspace settings to the colour profile of the camera? Does the camera have a colour profile, given its a monochrome chip?

I've got some more experimenting to do and will report back with any findings or breakthroughs if I make them:)

Just thought I'd toss it out there.

(Mods - please move if in wrong forum/section)

[iceman]

Hi Matt

Can you post the files straight out of stacking/processing in registax?

Then we can all have a play at recombining and you can check colour balance compared with your own.

[matt]

Will do, Mike.

Is it OK to do some wavelet processing in Regi (reference and wavelets) or do you want them pretty much 'virgin' jpegs?

I've got a few things I'm going to try when I get some more data, including seeing what happens if I stack more frames of certain channels to produce my TIFFS... such as more blue and green to compensate for this horrid red cast.

Cheers.

[iceman]

Wavelet processing is fine.

Stacking more of a certain channel shouldn't make a difference to the colour cast - it's more likely to do with the histogram of the channel at the time of capture.

[matt]

Yes. I am suspecting this has a significant impact, although I try to get the histo for each channel looking the same during capture.

Can you elaborate on this point, Mike?

What should I be mindful of during this part of the process?

Thanks for your help.