Moon 62.5%, 8-panel mosaic, 183C

[turfpit]

An 8-panel mosaic captured 5th May 2025, around 21:45 (twilight), 62.5% illumination, ~45° elevation. Processed with Autostakkert!4, Microsoft Image Composite Editor and PixInsight.

moon_8-panel_mosaic_183C_scaled.jpg (945.75 KiB) Viewed 1736 times

Hardware: Celestron C8 SCT, Celestron AVX mount, Altair 183C, UV/IR-cut/L filter in manual filter wheel, JMI motorised focuser.
Software: SharpCap 4.1, Autostakkert 4, Microsoft Image Composite Editor, PixInsight.
Capture: each panel was SER, RAW8, black_level=0, gain=200, exposure=18.5ms, duration 114s, frames=500 @ 4fps

2025-05-05-2044_0-Moon1.CameraSettings.txt

(1.36 KiB) Downloaded 47 times

During the session, there were 3 sets of captures:

• An 8-panel lunar mosaic captured with an Altair 183C

• A 7-panel mosaic captured with an Altair 183C

• A 7-panel lunar mosaic captured with an Altair 183M

The session had a duration of 1.5 hours and the data captured totalled 200Gb. 2 data sets were taken with the 183C as a contingency. The 183M was experimental.

Stacking
55% of frames were stacked (that is frames > 50% quality). The Alignment Point size was set at 200 which gave ~650 points. Drizzle was off (not needed with large capture area). All 8 files (each 9Gb) were dragged and dropped onto the AS!4 Open button and allowed to run unattended after the 1st Analyse.

Each output file (TIF) was 107Mb in size.

Stitching
The 8 TIF files were dragged and dropped into Microsoft Image Composite Editor, Stitch and Crop (with Auto Completion) applied and the result saved as a TIFF file. The output file was 260Mb with dimensions 8481x9003.

ICE_1.JPG (35.28 KiB) Viewed 1736 times

ICE_2.JPG (57.81 KiB) Viewed 1736 times

Post Processing
The order of processing was as follows:

• ColourCalibration

• Create RangeMask to protect very bright areas when stretching.

• CurvesTransformation for brightening and contrast (with mask applied).

• MultiscaleMedianTransform to sharpen (mask removed). Two applications with conservative values.

• CurvesTransformation, apply gentle saturation to taste.

Colour Calibration

ColourCalibration.JPG (92.65 KiB) Viewed 1736 times

colour_balance.JPG (73.58 KiB) Viewed 1736 times

After Colour calibration

after_ColourCalibration.JPG (38.1 KiB) Viewed 1736 times

Range Mask

RangeMask.JPG (61.93 KiB) Viewed 1736 times

Sharpening

MultiscaleMedianTransform.JPG (44.38 KiB) Viewed 1736 times

This is the resulting image from the mono camera:

moon_7-panel_mosaic_183M.jpg (750.06 KiB) Viewed 1736 times

2025-05-05-2138_4-Moon3.CameraSettings.txt

(1.39 KiB) Downloaded 38 times

Dave

[timh]

That's great. Again with all the details. I particularly appreciate the details of processing within PI and the use of the range masking. Now I think the problem that you have set is that it looks just about impossible to improve upon?

[turfpit]

Thanks Tim. Only took 8 years to get to here - I had this equipment in 2018. I reckon there is always room for improvement - better skills/knowledge, better software, better equipment.

Dave

[admin]

Hi Dave,

thanks for sharing - an excellent image (I bet it looks even better in full resolution!). I agree with Tim that it's helpful that you have shared your processing workflow - personally I rarely get as far as processing this sort of thing fully - once SharpCap has done it's job right, I typically move on to trying out something else...

BTW, did you capture the mosaic manually or use the mosaic tool in SharpCap?

cheers,

Robin

[Leandro Salgueiro]

Dave Excellent

[turfpit]

Thanks Leandro.

Thanks Robin.

did you capture the mosaic manually or use the mosaic tool in SharpCap?

This was a manual mosaic using the tried and tested crater overlap method. I didn't want to waste the high elevation of the May moon using new techniques. I am about to enter a period of 8 weeks of no astro darkness. During the 8 week period, I plan to have a look at the mosaic tool. As an ex sysadmin, I always take a pessimistic view of 'new things'.

Dave

[turfpit]

Here are some close-up views of the result when the 8 AS!4 panels (_conv files) were stitched and deconvolution applied. Details of the steps are below the images.

part1.JPG (877.13 KiB) Viewed 1393 times

part2.jpg (982.22 KiB) Viewed 1393 times

part3.jpg (974.84 KiB) Viewed 1393 times

Autostakkert!4 produced 8 sharpened panels (_conv in the filename). This was the post process.

• Stitch 8 _conv panels in Microsoft Ice

• Apply Deconvolution in PixInsight

• CurvesTransformation in PixInsight to apply gentle saturation 5 times

• CurvesTransformation in PixInsight to brighten the image

• Export as a TIFF giving a 258MB file

Because of the Autostakkert applied sharpening (_conv), no sharpening was applied in PixInsight.

These are the Deconvolution settings used (all defaults) with 10 iterations.

deconvolution.JPG (54.95 KiB) Viewed 1393 times

The TIFF file was loaded into GIMP and displayed at 25%.

• A very gentle Curves was applied

• Unsharp Mask (default settings) were applied

• The top third of the image was cropped, scaled to 75% and exported with 75% compression as a JPEG (to meet the forum 1Mb limit)

• The above was repeated for the middle and lower parts of the image.

The whole of the processing has been carried out using default settings or very small changes to settings - no brutal settings used here.

The JPEG compression has lost some of the detail. I have no experience of printing large images and wonder how much detail could be retained when printing the 258Mb TIFF image?

Dave