Struggling with Corona Australis - Would Appreciate Your Expertise! [Deep Sky] Processing techniques · Rod Prazeres Astrophotography · ... · 2 · 355 · 0

Hello everyone,I hope this message finds you well.

As an amateur astrophotographer who ventured into this mesmerizing realm a few months ago, I've been on an enriching journey filled with steep learning curves. The expertise and insights shared on this forum have been invaluable to me. If you're curious about my progress, you can check out some of my works on my astrobin here

My recent acquisition of the Redcat 51 has been exciting, especially transitioning from my 9.25 SCT (which I still use). Processing Corona Australis (NGC6729) has proven to be elusive. The core issue lies in rendering the clouds in the NGC6729 image.
For enthusiasts up for some detective work, I've compiled all the darks, flats, bias, dark flats, and raw images for this target. Please find them here.

A noteworthy point is this is the first time I'm using Dark Flats - could this be the core of the problem? 

Equipment and acquisition Details:

• 42x B x 180s
45x G x 180s
56x R x 180s
52x L x 180s
ASI160MM-Pro, Redcat 51, Bortle 6 sky

• Processing Software: DSS & Photoshop supplemented with GradientXTerminator, StarXTerminator, NoiseXTerminator

Processing Workflow:

1. Register each filter in DSS, incorporating the relevant flats, darks, bias, and dark flats.
2. Star detection threshold at 10%, eschewing noise reduction via median filter.
3. Utilizing the frame with the highest score as a reference.
4. Iterative registration for each channel with the specified reference frame (Note: I've experimented with and without this reference frame).
5. Transition of all master images for each channel into Photoshop, conversion from 32-bit to 16-bit, and setting to RGB mode.
6. Channel stacking and levels adjustment.
7. Photoshop stretching (achieved through manipulation of the Levels histogram sliders), followed by StarXTerminator, GradientXTerminator, and NoiseXTerminator applications. Fine-tuning involves multiple levels and curve adjustments.

DSS Stacking Settings:

• Result Mode: Standard (also tested Mosaic mode)
• Drizzling: Evaluated both standard and 2x drizzle.
• Alignment: Automated
• RGB Channels: Alignment off.
• SIMD Vectorization: Enabled
• Stacking Modes:
Light: Median KS (also tried Average)
Dark: Median KS (with hot pixel detection and removal). Also tried Median.
Flat & Bias: Median KS (also tried Median)
90% of images retained for stacking.

 I've delved into varied post-processing tactics, encompassing light pollution and gradient removal, alongside noise mitigation. Given the persisting issues, I'm inclined to believe a hiccup might have occurred during the stacking phase.

 While I acknowledge the prowess of PixInsight and plan to transition to it eventually, I'd love to unravel the potential of Photoshop for the time being, given its availability to me.

 Your time, expertise, and feedback mean the world to me, and I hope I'll be able to navigate this challenge.

 Warm regards,

Rod

Something to note too, don't stretch before starX, and GradientX. That limits the lower end of the star size, and can cause overshoots & undershoots in the background extraction from GX.

This only applies to PixInsight versions. The photoshop/affinity plugin version of SX requires some stretching first.

To quote Russell Croman in a support thread:

I definitely recommend performing the bulk of any planned stretching before using StarXTerminator in Photoshop or Affinity Photo. The PixInsight version is optimized for processing linear data, but the Photoshop version is not.