27/3 - Montes Caucasus LR deconvolution attempt.

[Blazar]

Not really played around too much with Astra image, but thought I would give it a go. After taking some images on the 27th, the x2 barlow attempts were not really cutting it using the standard processing. Michael Wilkinson advised to try AI3 and in particular L-R deconvolution instead of wavelets. So thought I would post this up (without my usual blurb) to see what people think.... using it has not introduced any additional harshness from wavelets, yet all I used were the default values. I am sure that as with any processing, pushing too far will kill it. I think this was the cauchy version of the sharpening. Any thoughts / feedback welcomed :-)

Thanks for looking as always.

[Astro Imp]

Hi Steve, I'm loath to criticise any of your images but to my eyes this does not appear as sharp as your usual offerings. This shows up especially in the lower part of the image around crater Callipus. 

Just my humble opinion.

All the best.

[Blazar]

No worries Alan, Thats why I posted it. I don't usually like to put stuff up that I don't think are good enough. Though this time I wanted other opinions ;-)  The barlow might not have been the best, and if its showing in one corner, then there may be an issue with it...or it just too much for the conditions. 

[Cath]

Just having a close look at the image Steve. I was just wondering if the 'tram lines' are real features or are they features introduced by the L-R deconvolution I wonder? .. have highlighted one area with the coloured arrow.

[cgarry]

I have never been sold on the idea that L-R deconvolution is better than well used wavelets.  This image backs up my opinion that wavelets in the right hands are superior.

Cheers,

Chris

[Cath]

oh the dual side by side image is not what Steve posted Chris, Steves original deconvolution image is on the left, the right hand image is the same image but with a small amount of processing I did to bring out some detail I was wondering about. I haven't used deconvolution or wavelets on it.

[cgarry]

oh the dual side by side image is not what Steve posted Chris, Steves original deconvolution image is on the left, the right hand image is the same image but with a small amount of processing I did to bring out some detail I was wondering about. I haven't used deconvolution or wavelets on it.

Yes I realised this!  I based my comments on the fact the image in the OP did not seem to be as good as Steve's usual standard.

The tram lines you pointed out are common for bright lunar crater rims, especially in sharp images.  I believe the general consensus is that they are caused by a diffraction effect and this is present in the raw data but brought out with sharpening. 

Cheers,

Chris

[Blazar]

Thanks Cath and Chris. The whole point was to put up an image that when I did use wavelets looked ok, but thought what would L-R dconv do for it. I think it came out a little soft. I guess thats down to seeing and the barlow. However concerning the tram line artefacts....I think this might be down to stacking...not too sure but I do see these when doing the close up work. All the caterlets seem to have an inner ring to them...I must admit Chris to not using PIPP (as I am waiting for the OS X version ;-) could not get it to run under WINE for some reason). could it be a misaligned frame somewhere in the stack? 

[cgarry]

I used to think the artefact was caused a misaligned frame, but gave up on that explanation because processing with AS!2, Registax or AviStack all produces identical artefacts.

Chris