A couple of May 2025 captures which have been processed using BiggSky blind deconvolution software (details below).
7-panel lunar mosaic captured 5th May 2025 with 60% waxing moon at 45° elevation.
Endymion/Atlas/Hercules captured 2nd May 2025 with 30% waxing moon at 55° elevation.
7-panel Mosaic
Hardware: Celestron C8 SCT, Celestron AVX mount, Altair 183M, UV/IR-cut/L filter in manual filter wheel, JMI motorised focuser.
Software: SharpCap 4.1, Autostakkert 4, Microsoft Image Composite Editor, BiggSky (deconvolution) and GIMP.
Capture: SER, MONO8, black_level=0, gain=250, exposure=5ms, capture area 5440x3648 , frames=500 @ 5fps
Endymion/Atlas/Hercules
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, BiggSky (deconvolution).
Capture: SER, RAW8, black_level=0, gain=200, exposure=31.5ms, capture area 1024x768 (to increase fps), frames=2000 @ 31fps
Processing Mosaic
Autostakkert!4: (for each panel) stack best 50% of 500 frames, AP size=200, no drizzle or resampling or sharpening (because that is what the documentation recommends).
BiggSky: All 7 panels loaded, deconvolution (PSF from ROI), not tiled, no denoise, exported to TIFF.
Microsoft Image Composite Editor: stitch 7 panels.
GIMP: rotate 90°, Levels, Noise Reduction (amount=1), Unsharp Mask (amount=500), Contrast=10, Curves to lighten, scaled to 40% to meet forum file size limit.
I elected not to apply denosie in BiggSky, as being a mosaic the image gets compressed anyway.
Processing Endymion
Autostakkert!4: stack best 50% of 2000 frames, AP size=200, no drizzle or resampling or sharpening (because that is what the BiggSky documentation recommends).
BiggSky: Deconvolution (PSF from ROI), tiled, 100% denoise before and after, crop, exported to TIFF.
** No further processing on the BiggSky TIFF output.
[Note: deconvolution runtimes - tiled=36s, non-tiled=22s]
BiggSky can perform blind deconvolution and requires the Matlab runtime library to be installed in order to run. The software is in the beta stage of development and can be downloaded at https://biggsky.com/. Documentation can be found at https://biggsky.com/ under the LEARN tab. The Matlab Runtime is downloaded and installed as part of BiggSky installation process. Once installed, a 30-day evaluation license can be requested via the application.
The software can process lunar, solar, planetary and deep sky objects and provides functionality such as blind deconvolution, denoise, crop, stacking, calibration, image analysis and project management.
BiggSky processing the 7 lunar panels looks like this:
I have only been in possession of BiggSky for 3 days and am working my way through the functionality. The Endymion image above was stacked in Autostakkert, deconvolved and cropped in BiggSky and had no further processing.
Higher resolution images of the above can be found at my astrobin account https://www.astrobin.com/wn690l/ and https://www.astrobin.com/az9cj6/
There are some threads on Cloudy Nights where people are sharing their experiences with BiggSky.
I hope you find this post useful.
Dave
Lunar with BiggSky blind deconvolution
Forum rules
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.
Please share the equipment used and if possible camera settings to help others.
Please upload large images to photo sharing sites (flickr, etc) rather than trying to upload them as forum attachments.
Please share the equipment used and if possible camera settings to help others.
Re: Lunar with BiggSky blind deconvolution
Amazing work, Dave!
Re: Lunar with BiggSky blind deconvolution
Thanks Brian. I am going for a lie down now.
Dave
Dave
Re: Lunar with BiggSky blind deconvolution
Dave,
Great images. Looks like a one stop bit of software for lunar images? Tim
Great images. Looks like a one stop bit of software for lunar images? Tim
Re: Lunar with BiggSky blind deconvolution
Thanks Tim.
BiggSky can stack SER files - next thing on my list to try. I can use the 3 craters file to test but I really need to capture some 'small data' for testing purposes. Mosaic panels still need to be stitched and GIMP is useful for 'final touches'. I still have a long way to go - it has been a hectic 3 days.
Dave
BiggSky can stack SER files - next thing on my list to try. I can use the 3 craters file to test but I really need to capture some 'small data' for testing purposes. Mosaic panels still need to be stitched and GIMP is useful for 'final touches'. I still have a long way to go - it has been a hectic 3 days.
Dave
Re: Lunar with BiggSky blind deconvolution
Tim
Here you go:
This is the TIFF file.
Still experimenting as the image can look too smooth with an aggressive denoise setting.
Multiple SER files could be loaded (I have tested this with multiple frames).
Dave
Here you go:
- Load 2000 frame SER, 1.5Gb, file into BiggSky
- Stack
- Deconvolve (PSF from ROI) with Denoise
- Crop
- Export TIFF and JPG
This is the TIFF file.
Still experimenting as the image can look too smooth with an aggressive denoise setting.
Multiple SER files could be loaded (I have tested this with multiple frames).
Dave
-
- Posts: 174
- Joined: Sun Jan 27, 2019 9:48 am
Re: Lunar with BiggSky blind deconvolution
Interesting.
Though it's not a free package.
MacOs version available on request.....!
No one appears to have used it on Solar images.....mmmm
Though it's not a free package.
MacOs version available on request.....!
No one appears to have used it on Solar images.....mmmm
Re: Lunar with BiggSky blind deconvolution
I agree - it is certainly interesting software. May well buy it if not too pricey, I wonder what it is doing exactly with deconvolution ? At least for deep sky deconvolution you can measure and parameterize the local PSF using stars as the point sources. Whereas as the heading implies -- for an ROI on the lunar surface -- you would have thought that it is blind in this regard?
PS. Actually - forget my question - I see David Biggar provides a very lucid summary of deconvolution and whta his software does on his site https://biggsky.com/faq/
A neural net type solution with the various vistas of the moon and planets as the (overly narrow) training set really would be cheating I think and this isn't that.
Tim
PS. Actually - forget my question - I see David Biggar provides a very lucid summary of deconvolution and whta his software does on his site https://biggsky.com/faq/
A neural net type solution with the various vistas of the moon and planets as the (overly narrow) training set really would be cheating I think and this isn't that.
Tim
-
- Posts: 174
- Joined: Sun Jan 27, 2019 9:48 am
Re: Lunar with BiggSky blind deconvolution
Hi Tim,
FAQs suggest Solar, but not Features.
No where can I find a price to justify the 30 free trial!
FAQs suggest Solar, but not Features.
No where can I find a price to justify the 30 free trial!