Jump to content

Noise - lens correction pattern?


jpk

Recommended Posts

Advertisement (gone after registration)

I have a very strange grid in files with noise which becomes more visible if I apply heavy noise reduction (see attachement). I guess this is the result from the in-camera lens correction... Is this true? If yes: is it possible to switch off this kind of processing without loosing the lens info in the EXIFs?

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

I guess this is the result from the in-camera lens correction... Is this true?

No, I don’t think so. The corrections applied in-camera wouldn’t have this effect. (Which camera, btw? And do you see this in JPEGs or DNGs?)

Link to post
Share on other sites

No, I don’t think so. The corrections applied in-camera wouldn’t have this effect. (Which camera, btw? And do you see this in JPEGs or DNGs?)

 

DNGs from the Monochrom, processed in LR5!

Link to post
Share on other sites

I turned off lens correction in the MM and took some test shots with ISO 8000. In LR4.4 and LR5 the pictures did not show any grid pattern (DNGs without any sharpening / noise-reduction). But after I rotated the pictures (+1,18) the grid appeared! And it's not only on screen, it's also visible in the exported TIFFs.

 

So a Lightroom bug?

Link to post
Share on other sites

The sample in my first post is a crop from this picture. But the problem exists in every picture with noise as soon as I rotate it with the crop / rotate tool.

 

The same hapens if I export a TIFF and rotate it in Photoshop - with or without prior noise reduction. So it seems it has to do with noise, but not with noise reduction...

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Link to post
Share on other sites

I have a niggling feeling of having seen this problem before as a LR bug.

It is clear that rotating an image in LR applies some extensive processing that does affect the image at a pixel level. See the post by LRuser24 in this discussion:

 

Lightroom 4.1 RC2: Artefacts when using shadows, clarity and image warp (lens corrections)

Link to post
Share on other sites

I agree with Jaap. Since the pattern appears only after rotation, then it is caused by rotation.

 

Incidentally, your correction of about 1.5 degrees is very much like my common error in framing. The error began to occur, perhaps by coincidence, when I switched to tri-focal glasses.

Link to post
Share on other sites

Thanks for your input!

 

I tried processing with and without NR, with NR before and after rotating, and also with different software (RAWtherapee) - it always shows the grid pattern after rotating. I even checked if files from other cameras show the same grid pattern after rotating - and they do!

 

The result is: files with heavy noise show grid patterns after rotating regardless which software is used. The process of rotating itself seems to introduce the grid patterns... Strange that I couldn't find any related information through Google!

 

The remainig question: what is the solution?

Link to post
Share on other sites

Did you try changing order of the processing steps? For example, if you apply noise reduction first and then rotate the image, rotate it first and apply noise reduction later?

 

I am afraid this is not possible, as Lr applies processing operators in a fixed order that makes sense (e.g. denoise first, then rotate).

Link to post
Share on other sites

The sample in my first post is a crop from this picture. But the problem exists in every picture with noise as soon as I rotate it with the crop / rotate tool.

 

The same hapens if I export a TIFF and rotate it in Photoshop - with or without prior noise reduction. So it seems it has to do with noise, but not with noise reduction...

 

Could you also post the same shot before rotation ?

Link to post
Share on other sites

I am afraid this is not possible, as Lr applies processing operators in a fixed order that makes sense (e.g. denoise first, then rotate).

 

Yes, but it is no problem in Photohop CS which uses the same algorithms.

Link to post
Share on other sites

Yes, but it is no problem in Photohop CS which uses the same algorithms.

 

Uhm... not so sure it is the same.

In any case, the OP says that the same happens in PS if he exports TIFF.

The only explanation I have is that the MM sensor noise patterns create some sort of Moiré interference with rotation algorithms.

Check also here, first picture is self explanatory:

http://en.wikipedia.org/wiki/Moir%C3%A9_pattern

Link to post
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...