Pulling night-shine from images of moons, how can i attempt night-shine extraction in Gimp on a RAW image? Options

[TrappistPlanets]

I tried to pull any possible nightshine from this umbriel RAW image (below), but i keep getting nothing, other than some terminator stuff, same goes for Triton, and Oberon (ik the night-shine for the 2 uranian moons would be low resolution, bs it was not the closes approach point like it was for Titania).


Also, is it possible to extract any higher detail plutoshine from RAW half phase Charon images (pluto is pretty bright!)?

i also want to try to pull out more detail of this stuff way beyond into the darkside of this image i found on one of the post encounter threads



so this is where i need help, how can i extract night-shine (either from atmosphere or from a nearby large object (like a planet)) in gimp?

[JRehling]

Umbriel's dark side would have "seen" a half Uranus, and thereby gotten light from Uranus roughly 1/500th that it got from the Sun. If a image from that geometry had enough bit depth that there was useful detail at 1/500th the contrast seen on the sunlit side, then you might pull out some detail. This already seems doubtful. Moreover, note that Voyager 2 was basically looking right past Uranus to see Umbriel, so there would have been no shadows cast by topography in the uranus-shine, and only albedo features would show up. On the known portions of Umbriel, albedo features exist but are atypical, and there's no guarantee that they exist in the dark side.

What you posted is an 8-bit image, so 1/500th of white is pure, uncontrasted black. Moreover, it's clear that there's light noise in the blackness of space around Umbriel, at levels of up to about 1/8th of the full range of brightness in the image. So, the noise is many times greater than any possible signal.

[mcaplinger]

What you posted is an 8-bit image...

And indeed, the raw out-of-the-camera images from Voyager ISS were 8-bit. So one should be looking for longer exposure and/or higher gain images (if there are any) to see dark stuff. As a general rule, if an image is properly exposed for the illuminated areas, it won't have enough dynamic range to see very far into the dark.

[TrappistPlanets]

And indeed, the raw out-of-the-camera images from Voyager ISS were 8-bit. So one should be looking for longer exposure and/or higher gain images (if there are any) to see dark stuff. As a general rule, if an image is properly exposed for the illuminated areas, it won't have enough dynamic range to see very far into the dark.

what about the Pluto (stuff way beyond the terminator in that one image i posted in the main post), Charon, and Oberon (thinking it would be the same case with Umbriel for Oberon, but Oberon has lots of albedo variation) i mentioned in the post?

[TrappistPlanets]

yeah.. i don't see anything regarding planetshine for umbriel
we either need ted, or there is truly nothing

[JRehling]

The Pluto image has detail for hundreds of pixels beyond the terminator, but at the level of exposure, that area of the image is speckled. It can look cleaner to downsample it, then turn up the brightness.

The problem with the areas of Pluto illuminated by twilight is that it's not completely straightforward what the detail means. With an airless body and a known point-like source of illumination, one can infer quite a bit. With illumination from an overhead sky whose luminance is itself not well characterized, it's in principle ambiguous whether, say, a dark patch in the image corresponds to darker albedo on the solid surface, slope due to topography, haze between the source of illumination and the surface, or haze between us and the surface. That's a lot of ambiguity.

[TrappistPlanets]

The Pluto image has detail for hundreds of pixels beyond the terminator, but at the level of exposure, that area of the image is speckled. It can look cleaner to downsample it, then turn up the brightness.

The problem with the areas of Pluto illuminated by twilight is that it's not completely straightforward what the detail means. With an airless body and a known point-like source of illumination, one can infer quite a bit. With illumination from an overhead sky whose luminance is itself not well characterized, it's in principle ambiguous whether, say, a dark patch in the image corresponds to darker albedo on the solid surface, slope due to topography, haze between the source of illumination and the surface, or haze between us and the surface. That's a lot of ambiguity.

so how can i bring out the detail in gimp?

[JRehling]

https://docs.gimp.org/2.10/en/gimp-tool-bri...s-contrast.html

[mcaplinger]

what about the Pluto...

The camera on NH was far better from the old Voyager camera, more bits and much better dynamic range.

[Phil Stooke]

Regarding the Uranian satellites illuminated by light reflected off the planet: you can only see it if you are looking at the part of the satellite which faces Uranus - if you are seeing the region facing away from the planet you won't see anything. In the absolutely ideal case you would see 50% of the moon lit by the sun and another 25% lit by the planet, but viewing directions will not always cooperate. You might like to see this blog post by Ted Stryk:

https://www.planetary.org/articles/1362


And this is the LPSC abstract that started it all:

https://www.lpi.usra.edu/meetings/lpsc2001/pdf/1074.pdf

Believe it or not, nobody had noticed this before. I can't believe it was over 20 years ago.

Phil

[TrappistPlanets]

Regarding the Uranian satellites illuminated by light reflected off the planet: you can only see it if you are looking at the part of the satellite which faces Uranus - if you are seeing the region facing away from the planet you won't see anything. In the absolutely ideal case you would see 50% of the moon lit by the sun and another 25% lit by the planet, but viewing directions will not always cooperate. You might like to see this blog post by Ted Stryk:

https://www.planetary.org/articles/1362


And this is the LPSC abstract that started it all:

https://www.lpi.usra.edu/meetings/lpsc2001/pdf/1074.pdf

Believe it or not, nobody had noticed this before. I can't believe it was over 20 years ago.

Phil

i down-sampled and messed with brightness and contrast, but i still can't see much detail other than some craters (marked in red in the pic with marks on it), and some faults (marked in yellow in the pic with marks on it)



how did how did ted remove all the instrument noise, that Pluto image is so noisy and makes it hard to see much detail in the night side stuff, other than some craters and faults/cracks

[JRehling]

how did how did ted remove all the instrument noise

The paper that Phil linked to discusses that. High pass filters accomplish what I suggested with downsampling.

If the relatively brief description doesn't make this clear, it was obviously a lot of hard work, not just trying one or two things.

The removal of instrument noise can often be served to some extent by removing the noise seen in dark frames, or, if you lack those, the dark areas of other frames. It is essential to find ones with the same capture parameters.

I think you're greatly underestimating the amount of difficult, persistent, autonomous effort that is required. This isn't going to be something where you skim the manual, try one thing, quote a link without reading what was in the link, and expect to be done. If you're asking other people how to do image processing, and expect to get a result that they haven't already gotten, you won't get anywhere. We all have image processing software. The careful, painstaking, laborious, autonomous effort is the thing in short supply, not access to image processing software.

On another note, circular dark spots in those areas need not be craters. Instrument artifacts are often circular, Pluto is known to have circular landforms that do not resemble impact craters, and as I noted earlier, the atmosphere itself introduces multiple sources of noise.

[TrappistPlanets]

The paper that Phil linked to discusses that. High pass filters accomplish what I suggested with downsampling.

If the relatively brief description doesn't make this clear, it was obviously a lot of hard work, not just trying one or two things.

The removal of instrument noise can often be served to some extent by removing the noise seen in dark frames, or, if you lack those, the dark areas of other frames. It is essential to find ones with the same capture parameters.

I think you're greatly underestimating the amount of difficult, persistent, autonomous effort that is required. This isn't going to be something where you skim the manual, try one thing, quote a link without reading what was in the link, and expect to be done. If you're asking other people how to do image processing, and expect to get a result that they haven't already gotten, you won't get anywhere. We all have image processing software. The careful, painstaking, laborious, autonomous effort is the thing in short supply, not access to image processing software.

On another note, circular dark spots in those areas need not be craters. Instrument artifacts are often circular, Pluto is known to have circular landforms that do not resemble impact craters, and as I noted earlier, the atmosphere itself introduces multiple sources of noise.

what about the possible cracks i marked in yellow, is that noise or am i onto something there

[nprev]

Topic closed.