work flow for SBIG-8300c

[wa4hnd]

I've been using the PixInsight trial now for a few days and so far I'm very impressed.  But there's only one thing I haven't been able to figure out...  Color conversion.  I must be missing a step cause when I do a Debayer ( RGGB ) I end up with a RGB image but there is no color in it... ie grey.

If I do the color convert first in MaxDL, then bring the subs over for registration and stacking, all is well.  I think I have all the basic basics under control except for this one step so I'm sure something is missing.

Any help would be  appreciated.

Al

Great piece of software ( or I guess I should say OS <g> )

[Nocturnal]

Hi,

could you either upload one of your debayered files or post a screenshot of the histogram of your debayered image? How did you verify the resulting image was monochrome?

[Philip de Louraille]

Are you sure the debayer matrix is RGGB?
http://www.kodak.com/global/plugins/acrobat/en/business/ISS/datasheet/fullframe/KAF-8300LongSpec.pdf

[Nocturnal]

Hi,

the debayer settings depend more on the driver and application than the sensor. They decide which pixel to read out first. This article describes an easy process for verifying the debayer settings. You can use PI for these steps.

http://qhy8.wikispaces.com/How+do+I+verify+my+colors+are+correct%3F

[Philip de Louraille]

If you have the time to put a raw (straight from the camera) FIT file on the net, I'll gladly look at it.

[wa4hnd]

Thanks for the replies.   I'll try and get a virgin file and upload it ( not exactly sure how or where to upload it to but I'll figure that out ). 

Ref RGGB, I heard someplace that was the pattern for the 8300... maxdl has a setting called SBIG in their convert menu so I set it to that and that produces the correct colors.... but it really doesn't tell me the pattern...  I'll try the link to figure it out.

But right now, I have an emergency as my frig is not friging and I'm helping wife carry food to frig number 2  <g>.   And as luck has it, looks like it will be clear tonight.

Thanks in advance for the help, I'm sure we'll figure this out.

Al

[wa4hnd]

Ok, I had time last night to grab a 3 min sub of the Lagoon Neb.  It was very low when I got to do it and going through a lot of smoke ( AZ fires ), but it does show color after a debayer in maxdl... but here is a copy of the raw fit before any process at all.

I posted a link to the .fit and also a zip of the fit.  For some reason the fit doesn't like to download ( need to brush up on my IIS I guess ) but the zip works.  You can find the links at http://ackerconsulting.com/astro


 Al

[Philip de Louraille]

GRBG is the Bayer matrix for PI for that orientation.

[Nocturnal]

Hi,

I also debayered it using GRBG and it has plenty of color. I'm not sure what the problem is. There are only 4 choices, surely you tried them all and saw colored images appear?

[wa4hnd]

Thanks,  I'll try GRBG again... I'm sure I did that but who knows at 4AM <g>   My plan is to use MaxDL for camera ops and pixinsight for process work... I just need to get the process down. 

So I'm thinking the flow should be...

Calibrate
debayer
register
stack
then the background tasks etc?

Excuse me for not knowing the correct PixInsight terms yet... I'll get there, I promise <g>

Correct or am I off base here?

Thanks in advance.

Al

[wa4hnd]

Ok, GRBG worked fine ( da )... sorry, must have been a late night. 

Now if someone can confirm my thinking on the correct workflow.. I'll get out of your hair for a few days. ( till my next newbie question ).

By the way I'd go for the I7, nvidia and as much ram as you can buy <g>.  I'm trying out PI with my laptop... intel core duo 2.6 thinkpad, running vista 64 and 8 GB ( the most I can get in that box ) and nvidia Quadro   and it seems to run great.

Al