Strange noise on the background, appears after processing raw images

Status
Not open for further replies.
I have noticed this in the background of a few images now where I had to push the shadows. What I discovered just the other day was that if I turn off lens corrections in Lightroom, the pattern disappeared. Strangely it seems to just be the distortion correction that does it; I would have thought it had something to do with vignette correction.
Flat field correction makes the pattern more obvious. Correcting vignetting should be more like flat field correction. Distortion correction could possibly affect the outcome.

If there is a gradient in the image then that affects the shape of the patterns.

Guess we will see more posts about this (non issue?) in the future. Lots of mishandled and pixel peeped files out there. :-D
 
I have noticed this in the background of a few images now where I had to push the shadows. What I discovered just the other day was that if I turn off lens corrections in Lightroom, the pattern disappeared. Strangely it seems to just be the distortion correction that does it; I would have thought it had something to do with vignette correction.
Flat field correction makes the pattern more obvious. Correcting vignetting should be more like flat field correction. Distortion correction could possibly affect the outcome.

If there is a gradient in the image then that affects the shape of the patterns.

Guess we will see more posts about this (non issue?) in the future. Lots of mishandled and pixel peeped files out there. :-D
Does not distortion correction replace one set of pixels with another. All pixels are remapped by a transformation. Pixel A (xy) is transformed to A (x1 y1).This is an interference pattern, which need two grids. Where are they?
 
I have noticed this in the background of a few images now where I had to push the shadows. What I discovered just the other day was that if I turn off lens corrections in Lightroom, the pattern disappeared. Strangely it seems to just be the distortion correction that does it; I would have thought it had something to do with vignette correction.
Flat field correction makes the pattern more obvious. Correcting vignetting should be more like flat field correction. Distortion correction could possibly affect the outcome.

If there is a gradient in the image then that affects the shape of the patterns.

Guess we will see more posts about this (non issue?) in the future. Lots of mishandled and pixel peeped files out there. :-D
Does not distortion correction replace one set of pixels with another. All pixels are remapped by a transformation. Pixel A (xy) is transformed to A (x1 y1).This is an interference pattern, which need two grids. Where are they?
If this is interference in the backlit image sensor (one of several claims) then the interference pattern should be baked into the file. More like interference in oil on water and soap bubbles (the image sensor thickness is fixed but illumination is not).

If so distortion correction should affect the pattern.

Have not yet found a good explaination regarding the colored patterns. Chip designers, opticians and image processing experts probably know what is really going on.
 
I have noticed this in the background of a few images now where I had to push the shadows. What I discovered just the other day was that if I turn off lens corrections in Lightroom, the pattern disappeared. Strangely it seems to just be the distortion correction that does it; I would have thought it had something to do with vignette correction.
Flat field correction makes the pattern more obvious. Correcting vignetting should be more like flat field correction. Distortion correction could possibly affect the outcome.

If there is a gradient in the image then that affects the shape of the patterns.

Guess we will see more posts about this (non issue?) in the future. Lots of mishandled and pixel peeped files out there. :-D
Does not distortion correction replace one set of pixels with another. All pixels are remapped by a transformation. Pixel A (xy) is transformed to A (x1 y1).This is an interference pattern, which need two grids. Where are they?
If this is interference in the backlit image sensor (one of several claims) then the interference pattern should be baked into the file. More like interference in oil on water and soap bubbles (the image sensor thickness is fixed but illumination is not).

If so distortion correction should affect the pattern.

Have not yet found a good explaination regarding the colored patterns. Chip designers, opticians and image processing experts probably know what is really going on.
Well do we know if it is baked in? Also DPP knowing these cameras, might be able to remove it as it has a moire control.
 
It is not moire, that pattern is from the lens profile that was applied. De-select it, and you'll see the pattern go away.

I don't think that the lens profile correction is critical for images like this, as any peripheral shading correction can be applied using the generic slider, and it is not an architectural-type image where distortion needs to be corrected. It always reveals itself in opened shadows.
 
18c6e24a9e3640b3af61464605404aed.jpg

Hi. Can anybody explain to me what are the geometric shapes in the background an how to avid them Thanks
If you get a chance, can you post a link to the RAW?

--
Joe
 
It is not moire, that pattern is from the lens profile that was applied. De-select it, and you'll see the pattern go away.
It's still a very typical moire pattern, even if it's caused purely by digital corrections - and most likely yes it's lens profile correction.
 
Here's an example of what the LR lens profile does to an image. This is image 19/156 by Dale Baskin from the DPR R5 sample gallery.

Here's the original image.

af0c940659c2400bb6f112286dbc9c5c.jpg

Here is the image with the LR lens profile turned on. I've pushed exposure and shadows way too far so that the pattern can be seen more easily in the field and tree area.

8450556cedee43b29046a2dee3b81eb5.jpg

Here's the same image with the lens profile turned off.

051742fac9174ce4bece8e7e04d56b23.jpg

The pattern differs with lenses and maybe other factors as well.

--
Joe
 
So these are digital artefacts caused two by 2 over laid grids interfering, because that is as far as I know, is the only way to cause moire.
 
Last edited:
So these are digital artefacts caused two by 2 over laid grids interfering, because that is as far as I know, is the only way to cause moire.

Can someone actually show this?
I just demonstrated what LR does in the post immediately above your post above.
 
I have old lightroom 6. Can I reproduce this with any of my noisy images and a lens profile? Is it a specific lens profile?
 
18c6e24a9e3640b3af61464605404aed.jpg

Hi. Can anybody explain to me what are the geometric shapes in the background an how to avid them Thanks
It would be good for you and everyone else to get to the bottom of this issue as you can tell from the responses that there have been MANY problems with Adobe Lightroom RAW conversion. Sometimes it is not Adobe's raw converter that is not the best and sometimes awful.

I would first try the free Canon DPP4 and see what you get.

Next, I would download DxO PL5 (free for 30 days) and try it. DxO PL5 (or PL4) is widely regarded as the best RAW converter, particularly in cases of higher ISO and recovering dark areas. It is also superior in dealing with Morrie (patterns in the real world causing lower frequency patterns in the image).

If you give a link to a raw file, people would be happy to see if it is a RAW conversion problem (I wish DP review had a repository for, say, a few RAWs at a time per person of RAW files).
 
I have old lightroom 6. Can I reproduce this with any of my noisy images and a lens profile?
I don’t know what you can do with LR 6. I use the current versions of ACR and PS.
Is it a specific lens profile?
This becomes visible when you try and push underexposed parts of a file too far in ACR or LR with the lens profile turned on.

I haven’t tested every lens, but I don’t think that it is lens dependent.

--
Joe
 
Last edited:
Does it depend on the number of pixels the camera has?

I have tried it on a 21mp image with a lot of black and pushed that, with exposure and shadows. I applied a profile. All I see is that with the profile the pixels are shifted as they should be . I see no grids apart from the noise of the sensor. I see no resulting moire.
 
I have old lightroom 6. Can I reproduce this with any of my noisy images and a lens profile?
I don’t know what you can do with LR 6. I use the current versions of ACR and PS.
Is it a specific lens profile?
This becomes visible when you try and push underexposed parts of a file too far in ACR or LR with the lens profile turned on.

I haven’t tested every lens, but I don’t think that it is lens dependent.
Well, I guess it's lens dependent to the extent that the lens profiles are lens dependent. So, I would expect it to be worse with lenses that need more correction.
 
Does not distortion correction replace one set of pixels with another. All pixels are remapped by a transformation. Pixel A (xy) is transformed to A (x1 y1).This is an interference pattern, which need two grids. Where are they?
As above, the Bayer pattern and the lens profile map.
 
18c6e24a9e3640b3af61464605404aed.jpg

Hi. Can anybody explain to me what are the geometric shapes in the background an how to avid them Thanks
It would be good for you and everyone else to get to the bottom of this issue as you can tell from the responses that there have been MANY problems with Adobe Lightroom RAW conversion. Sometimes it is not Adobe's raw converter that is not the best and sometimes awful.

I would first try the free Canon DPP4 and see what you get.

Next, I would download DxO PL5 (free for 30 days) and try it. DxO PL5 (or PL4) is widely regarded as the best RAW converter, particularly in cases of higher ISO and recovering dark areas. It is also superior in dealing with Morrie (patterns in the real world causing lower frequency patterns in the image).

If you give a link to a raw file, people would be happy to see if it is a RAW conversion problem (I wish DP review had a repository for, say, a few RAWs at a time per person of RAW files).
Also as suggested in the link, try debayering in DPP first, then export the TIFF to LR. This will show if it's a compatibility problem between the LR debayer algorithm and lens profile correction.
 
Does it depend on the number of pixels the camera has?

I have tried it on a 21mp image with a lot of black and pushed that, with exposure and shadows. I applied a profile. All I see is that with the profile the pixels are shifted as they should be . I see no grids apart from the noise of the sensor. I see no resulting moire.
No Mike. The effect doesn't seem to be dependent on the number of pixels.

Here's a screenshot of a sample image from the DPR low light sample gallery for the 12MP Sony A7sii that has been pushed too far. You can see the pattern the easiest in the top of the image.

a01f6e1c51a847da9da22ec354d6b6b6.jpg

I don't know why you are not able to reproduce the effect. If you want to post a link to the RAW image that you are trying to use to reproduce the effect, I'd be happy to look at it and see what I can do.

You've been posting with a lot of issues lately, and folks have asked you to post links of the RAW images that you are having trouble with in other threads. You haven't done it. I'd encourage you to post links to the RAWs when you are trying to solve a problem or have an issue. Without seeing the RAW files, it can be a lot of guesswork as to what your issues are and that doesn't move the ball forward.

--
Joe
 
Status
Not open for further replies.

Keyboard shortcuts

Back
Top