BlurXTerminator and oversampled data

[Adam J]

On 23/12/2022 at 10:33, ollypenrice said:

I think the one processed via BlurX is perceptibly better. What is it that you don't like?

Olly

Am with Vlaiv here the right one is more detailed and natural.

[ollypenrice]

11 hours ago, Adam J said:

Am with Vlaiv here the right one is more detailed and natural.

And there's the rub. We might want to make objective assessments of the results of this program but, even if we could, what would be the point? We are all going to observe the image subjectively once it's finished, so a subjective discussion is, in fact, the most logical one to have.

Olly

[Scott Badger]

17 hours ago, ollypenrice said:

I've been inputting PSFs generated by PSFImage but haven't yet come to any conclusions worth sharing.

Olly

I found with the embedded stars of NGC 206 in M 31, there was a lot of bridging with the default setting. Using a PSF diam of 3 px did a much better job, but that’s a much smaller diameter than the actual PSF of the image. I also played around with a lum integration of M51 and using a more accurate value (8 pix) significantly increased the de-blurring over either a lower inaccurate PSF diam or Auto, but a little unnatural looking. A sort of ‘glassy’ or ‘liquidy’ look, like I’ve seen with some of the examples posted. In the documentation, it’s noted that any coma etc. corrections to stars will also be applied to non-stellar features under the automatic setting, and it’s likely to overcorrect images or areas of images that don’t have stars. Not sure, though, how an inputted PSF works with a more localized application of deconvolution.

In any case, I think we still have much to learn about using BX, and the software probably still has some learning of its own to do, plus development tweaks and fixes, but seems like it’s being judged  against an expectation that v 1.0 is going to work perfectly on any image using the default settings.

I also understand those who feel it makes things too easy, but that’s a strictly personal opinion, and like woodworking, say, you can always choose a power tool or hand tool for any step of the process.

Cheers,

Scott

Edited December 31, 2022 by Scott Badger

[Scott Badger]

I’ll add that for anyone who hasn’t read the documentation for BX, I highly recommend it. It addresses nearly every concern that’s been raised and makes it clear that this isn’t magic. The tool’s output, like every other decon or sharpening tool, is an approximation. Many different sharpened images could be convolved into the same original and accuracy is dependent on many factors, including the settings used and the nature of the data inputted.

Cheers,

Scott

EDIT: RC's own words, "[...] it is up to each of us whether our final images are faithful representations of reality, or grotesque, over-processed messes. [....] A scalpel can be a precision life-saving tool, or a murder weapon, depending on how it's wielded."

Ha! I don't think any of BX's strongest critics, even, have used words like 'grotesque' and 'murder'!..... : )

Edited December 31, 2022 by Scott Badger

[ollypenrice]

Paul rebuilt our widefield M45 mosaic (linear data) after running them through BlurX at default settings. In doing the post processing I felt it held more delicate detail and the stars were easier still, using StarX.

Full size is enormous but here's a link to a fairly large one. Be sure to use the full screen icon, upper left, and click to enlarge the image. The partial screen sometimes creates artifacts.

https://ollypenrice.smugmug.com/Other/DUSTY-DARK-AND-MILKY-WAY-TARGETS/i-RXwsZdg/A

Olly, and Paul Kummer.

 

[geoflewis]

3 minutes ago, ollypenrice said:

Paul rebuilt our widefield M45 mosaic (linear data) after running them through BlurX at default settings. In doing the post processing I felt it held more delicate detail and the stars were easier still, using StarX.

Full size is enormous but here's a link to a fairly large one. Be sure to use the full screen icon, upper left, and click to enlarge the image. The partial screen sometimes creates artifacts.

https://ollypenrice.smugmug.com/Other/DUSTY-DARK-AND-MILKY-WAY-TARGETS/i-RXwsZdg/A

Olly, and Paul Kummer.

 

Wow....!! Just wow....!!

[teoria_del_big_bang]

1 minute ago, geoflewis said:

Wow....!! Just wow....!!

Pretty much all I could say too as I was almost speechless,
Magnificent Image. 
It's one of those images I see occasionally and keep a copy in a directory of a handful of images I have that is what I am aiming for as my imaging and processing improves.
In all probability I never will achieve such , partially due to my sky quality and imaging  time available, and some due to my own limitations and skills, but it always pays to strive towards something 🙂 

Steve

[tomato]

On 01/01/2023 at 16:35, teoria_del_big_bang said:

Pretty much all I could say too as I was almost speechless,
Magnificent Image. 
It's one of those images I see occasionally and keep a copy in a directory of a handful of images I have that is what I am aiming for as my imaging and processing improves.
In all probability I never will achieve such , partially due to my sky quality and imaging  time available, and some due to my own limitations and skills, but it always pays to strive towards something 🙂 

Steve

+1 for inspirational images. I still keep an image of Pickering’s Triangle posted on here by @gnomus several years back for the same reason.

[ollypenrice]

14 minutes ago, tomato said:

+1 for inspirational images. I still keep an image of Pickering’s Triangle posted on here by @gnomus several years back for the same reason.

A very good imager indeed.

Olly