Basic LRGB workflow part one
This tutorial describes a basic workflow to process LRGB data after calibration, alignment and integration. There are intermediate steps that refine the workflow however the objective of this tutorial is to provide a basic structure for imagers relatively new to PixInsight.
The basic steps are:
1. Create the RGB image with Channel Combination of Red, Green and Blue stacks
2. Dynamic Crop to remove stacking artefacts
3. Create a flat background sky free of gradients with Dynamic Background Extraction
4. Balance the RGB colours with Colour Calibration of RGB image
5. Stretching of RGB image with Masked Stretch
6. Histogram Transformation adjustments of non-linear stretched RGB image and removal of residual green gradient with SCNR
7. Stretching of Luminance image with Histogram Transformation
8. Preparatory mask work for combining Luminance with RGB
9. Channel Combination of Luminance with RGB to create LRGB image
10. Outline of post processing enhancements, blending and noise reduction
Step 1: Channel Combination of R, G & B images
I do assume a basic knowledge of PI terminology and techniques, for example how to rename the Image Identifier and to save a Project.
If you name the Image Identifier and end the name with '_ChannelColour' eg "Int_R" (which in my 'nomenclature stands for integrated red channel master', the channel option box will default to the respective colour channel, thus you save having to scroll down through the image windows.
Clicking the circular 'global' button will combine the three greyscale images into one combined RGB colour image.
Combined RGB Image
As an aid to myself, I rename the Image Identifier at critical stages to help track processing progress, which is invaluable when you reach final stages and you may have 20+ images minimised in the workspace.
Apply Screen Transfer Function to RGB
Opening the STF process, click on the tick icon on the lower right (this tracks the screen stretch for each open image on the workspace) and then the yellow and black circular symbol on the left, leaving un-ticked the diagonal linked chain icon. The linear image is now displayed without a permanent stretch.
Step 2: Dynamic Crop of RGB and saving the 'instance'
To make cropping of the image easier, I expand the image window a little. Opening Dynamic Crop, clicking the lower and furthest right icon, the 'reset' icon, places a white copr border around the image. You can then resize and/or rotate this border how you wish.
Before you apply the crop, drag the the New Instance icon onto the desktop to save the crop properties.
This saved crop can be applied to Luminance image (or any other open image windows). If you apply the crop before saving the New Instance, you will not be able to apply the crop to other images.
Apply New Instance of saved Crop to Luminance
It is vital that the same crop properties are applied to the Luminance to ensure that the image channels remain star aligned.
Step 3: Creating a flat background - Dynamic Background Extraction
I have found, more often than not, that using fewer data points manually sampled from the image edges and nebula/dust free regions is more successful than peppering the image with data points, whether manually chosen or generated using the Sample Generation options. Selecting points too close to a galaxy can also cause a dark ring around the structure and spiral arms.
In the M81 I am using for this tutorial I have selected 26 points.
If you are unsure where faint background dust or nebula may be lurking within the image you can use the Luminance image window with the Boosted STF as a template to select the sample points, saving the New Instance enabling you to apply it to the RGB or Luminance when you aready.
The PI documentation and Keller's Inside Pixinsight book provide explanations of the tool and also which options to change for specific problems.
In the main, the Target Image Correction wants to be set at 'Subtraction' and I have found that setting Tolerance to 0.750 works in many circumstances.
Occasionally I have had to use DBE twice and even sometimes on the non-linear stretched image.
You can of course blend the DBE image with the non-DBE image.
Saving the New Inastance allows the DBE properties to be applied to the Luminance image. Unlike Dynamic Crop, you can save the New Instance after it has been applied and retian its properties.
Background gradients after DBE
Applying the saved New Instance DBE to the Luminance
The Luminance image after DBE
As you start to accumulate image windows it is a good idea to minimise the windows and arrange them neatly to help keep track of your work or to easily return to an earlier stage. The minimised windows can be placed adjacent vertically in a chronological sequence or in one of the other workspaces.
The history of process steps is saved and I always save work periodically during processing in a project.