Clavius testing

[neil phillips]

I started wondering if i have been under sharpening my images. I do think there can be a very fine line. Detail retrieval is one thing. But when the leading edge of tiny craters starts to over whiten look stark. And the blacks start to posterize. I think balance has its place. The only thing that's sometimes hard to judge is where that balance should be. I wouldn't normally sharpen this much. Is it too much? Think i am becoming more uncertain?

Certainly, do not think there is any noise there. But there is different types of noise. something where the leading edge is starting slightly to over whiten. Can be seen as a type of noise too. Though technically it may not be as such. But it has the effect of looking slightly un natural. Or the best word perhaps is slightly over exaggerated. The only question is where does that point occur. Think it can sometimes be more difficult to judge than some might think. And  I am certainly revaluating my own judgment

 2016 12" Newtonian NEQ6 PRO ZWO 290mm IR685 Registax. AS/3 Astrosurface Pipp. Image analyzer. Seeing average. Soon to get my new EQ6 R Mount. In Dec  And hopefully a 12 or 14 " Newtonian running in the new year. Quite excited actually.  Good as my 10" is you can't beat resolution. Even under  average conditions  Really looking foward to getting bigger optics on the moon and planets again. I am sure its going to be a lot of fun. 

100% Capture size

 

 

 

Edited November 8, 2022 by neil phillips

[astrolulu]

I think that you have very accurately defined the not always perceived limit of acceptable image sharpening (but also enlarging!).

Apart from the noise, it is the effect of posterization and edge duplication that you mentioned, which makes the outlines of objects - such as craters - seem unnaturally thick, as if they were girdled with a bicycle tube. Of course, this effect is more pronounced the smaller the diameter of the telescope, so one of the escape routes seems to be simply escaping into larger diameters.

Recently, I have noticed that a good effect is achieved by enhancing the blacks in the shadows in such a way that at least some of the posterization falling on the dark parts disappears (or is less visible). The image seems more precise even though it was not sharpened:

Edited November 7, 2022 by astrolulu

[neil phillips]

13 hours ago, astrolulu said:

I think that you have very accurately defined the not always perceived limit of acceptable image sharpening (but also enlarging!).

Apart from the noise, it is the effect of posterization and edge duplication that you mentioned, which makes the outlines of objects - such as craters - seem unnaturally thick, as if they were girdled with a bicycle tube. Of course, this effect is more pronounced the smaller the diameter of the telescope, so one of the escape routes seems to be simply escaping into larger diameters.

Recently, I have noticed that a good effect is achieved by enhancing the blacks in the shadows in such a way that at least some of the posterization falling on the dark parts disappears (or is less visible). The image seems more precise even though it was not sharpened:

The image wasn't enlarged i don't think. it's been a while. I was considerably oversampled when i took the capture. so, the size was coming at capture.

Yes, I am aware of the black's improvement I often do it myself. But I also don't like the effect it can have in other ways. Especially when strong sharpening is used. It can also lead to posterization effects. And exaggerating strong sharpening in other ways. However, the tweak here doesn't look too bad. I originally posted 4 images up here with various sharpening and levels. But thought it too confusing. And I didn't like most of them. They were put up for comparison sakes. Leaving just this one capture here. that i preferred the sharpening on. 

So the large craters look better with darker blacks. However the small detail on the crater floor now looks more exaggerated. Often you dont get something for nothing. Beauty really can be in the eye of the beholder. Cheers Astrolulu 

My quick attempt at another tweak sorting out the craters 

 

Edited November 7, 2022 by neil phillips

[astrolulu]

You're right - the method of hiding part of the posterization simply has no chance to work where dark area is not big enough - that is, on small details. Cheers 🙂

[neil phillips]

5 minutes ago, astrolulu said:

You're right - the method of hiding part of the posterization simply has no chance to work where dark area is not big enough - that is, on small details. Cheers 🙂

Comparing these two recent clavius captures i think the 224 capture on the other post has come out far better. it has a much finer level of detail. I only realised this recently returning to the captures and playing with them again

[astrolulu]

I will mention one more phenomenon that caught my attention. Well, Photoshop uses several degrees of magnification of the image on the screen, between which you can easily switch using keyboard shortcuts.

I've noticed that whenever I bring a photo to a scale and sharpening degree that I might call "acceptable", going one notch lower on the screen magnification scale makes it appear much better, almost perfect. It is obvious that by reducing the image (it can be additionally sharpened) we get sharper "drawing and texture". The details themselves get smaller and partially disappear of course.

I am talking about this because a cost-free recipe is within easy reach - just reduce the image to make problems such as thickened edges, blur or noise disappear. Of course, the cost does exist - these are the details that we fight for 🙂

66.7% - which is the inverse of the magnification to 150%. This is the picture magnification level where everything seems perfect to me compared to the 1: 1 version ...

 

Edited November 7, 2022 by astrolulu

[neil phillips]

1 hour ago, astrolulu said:

I will mention one more phenomenon that caught my attention. Well, Photoshop uses several degrees of magnification of the image on the screen, between which you can easily switch using keyboard shortcuts.

I've noticed that whenever I bring a photo to a scale and sharpening degree that I might call "acceptable", going one notch lower on the screen magnification scale makes it appear much better, almost perfect. It is obvious that by reducing the image (it can be additionally sharpened) we get sharper "drawing and texture". The details themselves get smaller and partially disappear of course.

I am talking about this because a cost-free recipe is within easy reach - just reduce the image to make problems such as thickened edges, blur or noise disappear. Of course, the cost does exist - these are the details that we fight for 🙂

66.7% - which is the inverse of the magnification to 150%. This is the picture magnification level where everything seems perfect to me compared to the 1: 1 version ...

 

Again yes it does sharpen them up. I did downsize to 85% from capture. But did wonder if i should have gone further. I often do this especially with mosaics

The effect your describing seen here. massively downsized. But good to see the whole moon without too much panning. 

https://ibb.co/k4jvvwW

Edited November 7, 2022 by neil phillips

[astrolulu]

Your mosaic is razor sharp - and every picture should look like that. Unfortunately, this means that we have to say goodbye to some of the details that we fished so painstakingly from outer space 🙂

Recently, I noticed that in the case of Mars, the opposite effect - the brutal enlargement of the image above its original size - allows you to see details that are not visible on a 1: 1 scale ...