RGB combination and workflow

[shurik]

Hi,

I am a new user to this program and I have a question about RGB combination and processing. I am working on NGC7331 and collected decent Lum data, however due to the unstable weather and equipment guiding problems on the second night I ended up with 120 minutes of "R" (40x3 min.); 77 minutes of "G" (77x1 min.) and 130 minutes of "B" (130x1min.) . These images were shot at moderately polluted skies that produce red background.

1. How do I properly assign each channel weight during combination, or how do I calculate it? is there are any way to factor in light pollution that is mostly in "R" channel?

2. What is the proper work flow for RGB enhancements?

I would appreciate all advices, I spent a couple of weeks trying to come up with decent color but always end up with pale red or pale blue.

I uploaded 3 files (Red, Green, Blue) that are the product of calibration and registration, if someone has time to take a look at them and use their technique to come up with some results, I would greatly appreciate your help in posting your steps and final image.

files location:
http://www.mediafire.com/?gey19mj76d5g3

The collected data might be so bad it is not worth wasting time in this case I d like to know that too

thank you, Alex

[avastro]

Hello Alex,
Welcome to the Pixinsight forum.
Starting by Harry's video will help you in your learning curve.
http://www.harrysastroshed.com/pixinsighthomenewbie.html
As a general rule, average or bad data will complicate the process of your images.
Some targets are more difficult and should be planned later to be processed when your PI knowledge is appropriate.
About your questions;
1: I do that combining RGB channels using channelcombination tool without assigning weight to the channels (usually I collect the right number of sub exposures in relation with my filters and CCD chip characteristics), then I do a backgroundneutralization and or a colorcalibration.
2: I check for gradients, if needed I remove it using dynamicbackgroundextraction tool.
When I'm happy with my linear image I use the script maskedstrech in script utility  I do  around 140 iterations and 0.17 target median.
Then depending of the image, I do some improvement.
You probably should need to learn to do a proper star mask, to use a trouwaveletTransform,HDRWT, Pixelmath, ACDR...

Cheers


Antoine

[Andres.Pozo]

Hi Alex,

my workflow for RGB images is this:

• Calibrate the subframes using PixInsight or another tool.
• Choose the best/preferred subframe as a reference for registering.
• Register all the subframes using the same reference image for all channels. Doing this you don't need to register the integrated frames.
• Integrate the subframes for each channel
• Create an RGB image using ChannelCombination or PixelMath
• If the image has any gradients, apply DynamicBackgroundExtraction
• Calibrate the color using BackgroundNeutralization and ColorCalibration

After this there are several different paths that you can take depending on the scale and quality of the image and your own style of processing.