0.90
#...
·
 |
|---|
|
Hi I have a doubt about using BXT. I know that deconvolution (deconvolution tool) applies exclusively to luminance. As for BXT, should it also be applied to RGB? In Russ Cromann's tutorials I understand that it should be used on both. How do you use it? clear skies Riccardo |
7.90
#...
·
|
|---|
| Essentially, is up to you. I use it on both but not always. |
11.99
#...
·
|
|---|
andrea tasselli: Glad to see you are using it. I know you were one of the skeptics, originally  I also use it on both. |
|
0.90
#...
·
|
|---|
|
i also use it in both |
5.61
#...
·
|
|---|
| Typically on both but for some images where you might only use the RGB for the stars, there is no point. |
|
7.90
#...
·
|
|---|
Arun H: Touche'. ...after eating my ample serving of humble pie... Having said that, it does still introduce a fair amount of artifacts at times so I use it with a great deal of caution. |
0.00
#...
·
·
2
likes
|
|---|
| One of the best aspects of using BXT on RGB is that if there is any chromatic 'thing' going on w/ your stars... like maybe slightly different PSFs for your masters that make up RGB... BXT can correct those problems and even 'aberrations' if your system has them.... |
3.71
#...
·
·
1
like
|
|---|
|
Hi Riccardo, Like many of the above answers, I use BX and NX on both the Luminance and RGB composites, saving them as both native XISF and TIFF. I do my final touch-ups to the LRGB combination TIFF in Photoshop, and many times, I'll send the image back to PixInsight for another tweak with NX; it's interesting just how much extra 'smoothing' is done with this extra application. As far as OSC images, I typically extract the Luminance frame and process it separately, using both BX and NX. I'll do this for the image as a whole also, then recombine the Luminance with the whole image. The resulting increase in resolution is usually very good - sometimes, it's a bit much, sometimes not - the beauty is the choice. Be sure to experiment - that's a wonderful option, also. - - Steve |
8.52
#...
·
|
|---|
| I only run it in Luminance as my RGB data doesn't provide any detail, only color. I processing the stars independently of both stacks. |
0.90
#...
·
|
|---|
| I use for both. It really tightens up my stars, and it can help RGB. There is a lot you can do by playing with parameters or using masks too to get the effect you want. |
2.11
#...
·
·
1
like
|
|---|
|
It works extremely well with RGB data, and I don’t think there has been any sort of guidance from the developer otherwise. In fact, it has the capacity to correct some imbalance between the color channels in star models. I have also noticed some cases with my own work where it produces superior results when given the RGB data as opposed to working with a luminance extraction from the same image (possibly due to the fact that the modeling it does for deconvolution takes into account aberrations in stars and can extend some of that correction to the deconvolution applied to background details). The extent to which it can create issues with stars depends on their aberrations. If they fall outside BlurX’s training model, it can do janky things, so it needs to be used with regard to those limitations. You also have the option to work with stars (e.g. correction only can be valuable as well as reduction or halo adjustment if the star data can handle it) or focus on the likes of deep space object details. Very versatile program and it is pretty easy to make the most of it with regard to the data’s limitations. |
3.61
#...
·
|
|---|
| Like many here, I use it on both. I'd also be interested in hearing how many people routinely use it on non-linear images too. I've had successful results applying it in that way although it has to be carefully done or things go bananas very fast. |
1.81
#...
·
|
|---|
| I use on all the layers: L, RGB & SHO layers. I have not see any issues. I use with "correct first" option. |
4.02
#...
·
|
|---|
| Always BXT with any starless although it also does work good for star+starless images |
9.03
#...
·
|
|---|
|
I use on both. If you run the FWHMeccentricity script you may be shocked to see what it does to FWHM and star count. In my hands does not do much for eccentricity. Downside is that the stars are smaller, which I generally like, but it is a substantial change. Believe best to run all these AI variants on the stacked images as close as you can to the raw stack since the AI training is done on those images, I understand. |
