Moon - From color footage to a finally black-and-white image. What is the best way? - Forums

Topic starter

Feb 10, 2020

Dear moon guys.
I've been wondering for a long time what the best way in the image processing is to get the best black and white result. You have to know that I am using a ZWO ASI color camera. When is the best time in the editing process to convert an image into black and white? Before stacking or before sharpening and denoising. Is it better to edit the color channels completely separately and only merge them at the end? I don’t have any idea.
I am looking forward to your feedback.
Thank you in advance.
Axel

Engaging

astropical

Perfect post

Feb 11, 2020 · Edited: Feb 11, 2020

Hey Axel,

Having been in a similar situation with my ASI120MC-S color camera, I attached an IR-pass filter to the color camera and set white balance values WB_R and WB_B to 50 (no white balance). During processing in PS, I converted the image to grayscale as the first step because only then can best monitor what is happening to the grayscale image when you apply processing steps. There may be a wiser way for processing, but it worked for me.

This method does not match a mono camera for sensitivity while the IR-pass filter drops the camera's relative response down to some 50%. The shorter the IR pass wavelength the less loss in permeability. I think the Astronomik Pro-Planet IR742 is the best compromise between permeability and the filter's main purpose, namely oppression of poor seeing. My ZWO IR850 passes at 850mn and above. It is highly effective but way too 'dark'.

I have posted a graph on my website:
https://www.astropical.space/zwoasi290.php

Please expect further replies before you purchase anything :-)

LG, Robert

Helpful Supportive

Peter

Feb 13, 2022

I'm not a moon guy, but I've just taken 3 different photos from 3 different cameras on 3 different refractors. Each photo is 40-60% stacked in AS!3.x from a single 100 frame SER file - hard to pick a winner, and I don't really know what I'm doing. Used AS!3.1 for processing. SER player for preprocessing the colour captures as AS could not open the 16bit colour SER. The skies are noisy at the moment so I played around with different cameras doing different things. I am thinking the green bleed I see from colour cameras might be due to pixel resizing the goes on, and the final layering within the software. But that is only a guess as ASI cameras are strongest in green and it could be something to do with the debayering of the RGGB. I am guessing the green is significant somehow.

1. My guide camera is a Skywatcher 50 ED with a ASI290MM. Taken last night 12Feb2022.
Captured in 16bit mono, 100 frames; 1ms at 150 gain; AS!3.1, AP 60% frames, drizzle 1.5, greyscale color; sharpen; jpg from TIF (<1MB)
Straight through, zero editing, cropped and sized to under 1MB
(note: only single SER file lights were used in this testing, no other darks, flats or calibrations were used)

2. RedCAT 71 with AS178MC - I have a thread on this, could not get the green color bleed out of the final; processed in RGGB in AS!3.1 from 10Feb2022
Captured in 16bit colour, 100 frames; 1ms at 150 gain; AS!3.1, AP 40% frames, drizzle 1.5, align RGB; sharpen; jpg from TIF (<1MB) tried many different things but could not remove the green bleed completely
2a)

2b) the same image converted to greyscale (cheating) - 2B is probably my favourite so far.

3. 127ED APO with 0.7x FF/FR and ASI183MC pro at 0deg C - retained the RGB
Captured in 16bit colour, 100 frames; 1ms at 150 gain; AS!3.1, AP 66% frames, drizzle 1.5, align RGB; sharpen; jpg from TIF (<1MB) all in RGB colour
(note: when resizing this down I notice a green edge to some of the craters depending on bicubic sampling, as someone else mentioned this could likely be a sharpen artifact where pixel placement is critical, the green only comes out on sharpened images) - 3 looks nicer (softer) in full resolution; 2B looks cleaner in full resolution - all processed in a similar way

3B) is a cropped version of 3A) at 100% of the original
(note: the posting into Astrobin is viewing different side by side with what I have uploaded due to some processing by Astrobin; the same file looks crisper and cleaner on my monitor)

4. 127ED APO with 0.7x FF/FR and ASI462MC shot in 16bit mono
Captured in 16bit mono, 100 frames; 1ms at 150 gain; AS!3.1, AP 40% frames, drizzle 1.5, greyscale; sharpen; jpg from TIF (<1MB) all in mono
AS!3.1 upsizes the processing as this is only a 1936x1096 camera; this jpg from a bicubic resized TIF

I played around with AS13.1 settings under colour (grayscale if mono vs Forced RGGB if colour)
The colour processing benefited from Image Calibration column noise and row noise; but this didn't work for mono.
(note: only single SER file lights were used in the above, no other darks, flats or calibrations were used)

Peter

Feb 13, 2022

In my brief experience, it seems to be cleaner to retain the colour throughout the processing and convert to greyscale at the end. At least from what I have seen. But I am only guessing. The bit depth gives you more flexibility with grading in tones in the final image (it doesn't improve pixel accuracy) - this is common knowledge for photographers who chase higher bit depths (dynamic range) for shadows and highlights. Imaging the moon would benefit from higher bit depth captures due to the large dynamic range from very bright highlights to dark shadows. Also the files sizes are much larger ...

The same files as above; 3A and 3B now converted to greyscale (your end game). This makes 3A a clear winner for me, but it's hard to pic a favourite.

Helpful