Reading mode: Light Dark

Let's talk about moire removal

Go to last post
Bob, w/o any doubt, you master the English language much better than I do ;)

I did NOT assume line skipping is a moiré filter. What I said is that line skipping makes the effect of an AA filter almost vanish. And we're not talking about scaled images here.

The point is that standard Bayer demosaicing can deduce correct colors and avoid color moiré if color information is spread across a 2x2 Bayer cell such that the original RGB color information isn't completely lost. As is achieved by the original AA filter.

However, in video, there are normally too many pixels skipped in readout such that the color reconstruction algorithm fails even with an AA filter. An AA filter working at a lower spatial frequency would be required. And at such a lower spatial frequency as is typical for HD video readout, MTF for the D800 and D800E are equal.

However, I make an assumption here: That pixels readout in video mode are sparse and not blocked into groups. This remains to be seen until full tests can be performed.
bob elkind said:
falconeyes said:
I don't know all the details yet. But I expect the D800/E to do some line skipping in video.

This would mean that the AA filter is defunct in video which uses a larger pixel pitch.
Click to expand...
Issues on two counts:
  1. Your use of supposition speaks for the use of 'may' rather than 'is'. Let's not use supposition to infer certainty.
  2. Line skipping is as likely to exacerbate (amplify) artifacts which are present, rather than reduce or mask them. It rather depends on the line-skip (spatial) frequency vs. the artifact (spatial) frequency.
bob elkind said:
I would be surprised to learn that color moiré and line flickr issues in video would differ between the D800 and D800E.
Click to expand...
If video derived from an image with artifacts matches video derived from a similar image without artifacts, this means that the original (un-scaled) image doesn't matter. That would be very surprising.

--
Bob Elkind

Family, mostly sports. Seriously, folks, I'm not that good. If I can do it, you can do it!
photo galleries at http://eteam.zenfolio.com
my relationship with my camera is strictly photonic
Click to expand...
 
falconeyes said:
Steve Bingham said:
Please note that these do NOT work with video.
Click to expand...
I don't know all the details yet. But I expect the D800/E to do some line skipping in video. This would mean that the AA filter is defunct in video which uses a larger pixel pitch. I would be surprised to learn that color moiré and line flickr issues in video would differ between the D800 and D800E.

But the tips provided for fixing moiré in stills in some instances are very nice. Thanks!
Click to expand...
Hopefully, video will use other techniques - in camera! They have been dealing with this in $80,000 studio TV cameras for years. I bet they have a solution or two. We shall see.
--
Steve Bingham
http://www.dustylens.com
http://www.ghost-town-photography.com
 
tony field said:
Now that is MOIRE in spades.

Seems to me the techniques we have discussed here are based on RGB images, not B&W. Would be nice to try a fix based on a full colour image.
Click to expand...
Actually, that is an RGB image, but it might be a duotone, or a partially desaturated version. My point was that moire can be exceedingly hard to eliminate if it is very strong in the L channel. In that example, there is almost no moiré in the a or b channels.

For many (most?) of the moiré reduction tricks that I have seen, the effectiveness is best when the artifacts are visible as color fringes, distinct from background. And even then, these methods often leave a residual pattern in the L channel.

Thom Hogan, in a recent comment on his google+ stream, hinted that there may be some new algorithms emerging, which exploit properties of the Bayer mosaic. But he did not elaborate, saying that patents were involved. So maybe there's hope for the future. But I would not expect miracles. The theory suggests that some results of aliasing are impossible to separate from the original scene properties.
 
Nikon has today released View NX2 v.2.3.0 which can read D800 files.

I hope Nikon soon releases their new version of Capture NX2 with the promised moire reduction tool. With this in hand together with some D800E NEFs with moire we could make a much more educated decision.

Nikon please give us the new Capture NX2 soon together with some D800E NEFs displaying moire.
--
Kind regards
Kaj
http://www.pbase.com/kaj_e
WSSA member

It's about time we started to take photography seriously and treat it as a hobby.- Elliott Erwitt
 
Try boosting the saturation in PS. You would at least have a shot at it. Without color to work with, reducing color moire is tough. :) I had some luck doing this with the big tank. The one on the far right became impossible for me. In short, I pass. This might well take hours as opposed to minutes - and even then you might not succeed.
--
Steve Bingham
http://www.dustylens.com
http://www.ghost-town-photography.com
 
FYI - Capture NX2 ver. 2.3 has the moire reduction tool in it and is available for download from nikonusa.com
 
LarH said:
FYI - Capture NX2 ver. 2.3 has the moire reduction tool in it and is available for download from nikonusa.com
Click to expand...
I don't know how to interpret the statement on the D800E for instance on dpreview's preview: "It will be accompanied by an updated version of Nikon's Capture NX software, which will include a moiré reduction tool."

It could mean that a new version of capture NX2 will be released with the camera or it could mean that the present version of capture NX2 is include for the price of the D800E. or it could mean a new version of Capture NX2 with a new version of the moire reduction tool included with the D800E.

I guess we have to wait and see.

Anyhow we would need some NEFs from a D800E with moire to judge how well the moire reduction tool works on D800E NEFs. Moire reduction on NEFs should give better possibilities than working on a tiff or jpeg.The tool could partly work directly on the demosaicing.

--
Kind regards
Kaj
http://www.pbase.com/kaj_e
WSSA member

It's about time we started to take photography seriously and treat it as a hobby.- Elliott Erwitt
 
Kaj E said:
LarH said:
FYI - Capture NX2 ver. 2.3 has the moire reduction tool in it and is available for download from nikonusa.com
Click to expand...
I don't know how to interpret the statement on the D800E for instance on dpreview's preview: "It will be accompanied by an updated version of Nikon's Capture NX software, which will include a moiré reduction tool."
Click to expand...
It is confusing, I don't normaly use Capture NX so I downloaded the trial and found the moire reduction tool
Kaj E said:
It could mean that a new version of capture NX2 will be released with the camera or it could mean that the present version of capture NX2 is include for the price of the D800E. or it could mean a new version of Capture NX2 with a new version of the moire reduction tool included with the D800E.

I guess we have to wait and see.

Anyhow we would need some NEFs from a D800E with moire to judge how well the moire reduction tool works on D800E NEFs. Moire reduction on NEFs should give better possibilities than working on a tiff or jpeg.The tool could partly work directly on the demosaicing.
Click to expand...
I agreee it would be very nice to have some D800E NEF samples with moire so we could test it. - Larry
Kaj E said:
--
Kind regards
Kaj
http://www.pbase.com/kaj_e
WSSA member

It's about time we started to take photography seriously and treat it as a hobby.- Elliott Erwitt
Click to expand...
 
Digital display devices can also suffer Moire since they are themselves a sampling device.

In fact you can see Moire appear on a lower zoom of an that is not apparent on a higher zoom.

On certain interference patterns generated by small periodic patterns, our eyes will even see a Moire pattern. Moire is similar to the concept of a "beat frequency" in acoustics. This happens because of interference between two very close frequencies that causes a modulation on the envelope of the combined signal or a "beat."

Truman
David Nall said:
One of the images on the D800e picture examples showed moire, the one of the buildings.
http://mansurovs.com/nikon-d800e-image-samples

When I downloaded it, the moire was gone. I don't see any on the image of the Geisha girl pic either but someone else posted a pic showing the moire on her clothing. I am thinking that the problem might not be the same as it was on the D70. We will just have to wait and see as more pics become available. Dave
--
Visit my gallery at http://www.poperotzy.smugmug.com

View of Yosemite Valley, Half Dome and Glacier Point as seen from the Four Mile Trail.

Click to expand...
--
Truman
http://www.pbase.com/tprevatt
 
Steve Bingham said:
Rather than go through the theory (Marianne and others would be best for this) let me present a couple of ways of eliminating it using Photoshop. These really are not difficult techniques once you do them a time or two. Please note that these do NOT work with video.

Here is a quick tutorial (#2) but, for the sake of brevity, it only went so far. However, it is a must read for anyone considering the D800e that isn't already familiar with the many techniques . I prefer these two methods in this order. #2 is easiest, #1 is best . . . it just takes a little more knowledge of PS.

1- http://vimeo.com/23508129
2- http://www.youtube.com/watch?v=Ripf5Mcu2q4

--
Steve Bingham
http://www.dustylens.com
http://www.ghost-town-photography.com
Click to expand...
--
http://tourist-of-light.blogspot.com/

Please feel free to criticize, make suggestions, and edit any of my photos & re-post, to help show me 'the way'. * I am trying to Elevate the Level of my 'Snap Shots' :)
 
falconeyes said:
... What I said is that line skipping makes the effect of an AA filter almost vanish. And we're not talking about scaled images here.
Click to expand...
Line skipping removes data, including both artifacts and 'real' data. To the extent that 'real' data survives the line-skipping filter, artifacts will also survive the line-skipping filter. At the same time, line skipping can both introduce a new set of artifacts and exaggerate aliasing artifacts in the original image data.
falconeyes said:
The point is that standard Bayer demosaicing can deduce correct colors and avoid color moiré if color information is spread across a 2x2 Bayer cell such that the original RGB color information isn't completely lost.
Click to expand...
Please explain how your de-mosaic process works, and why it is immune to colour moiré. And please explain how you can deterministically discern 'original RGB color information' in an aliased image.

--
Bob Elkind

Family, mostly sports. Seriously, folks, I'm not that good. If I can do it, you can do it!
photo galleries at http://eteam.zenfolio.com
my relationship with my camera is strictly photonic
 
You must log in or register to reply here.

Similar threads

Steve Bingham
Let's talk drawing tablets
2
Replies
29
Views
30
yanshen
Y
Erik Kaffehr
Elements of image quality...
6 7 8
Replies
149
Views
3K
JimKasson
JimKasson
E
Vimeo- a waste of time and money!
Replies
11
Views
765
larsbc
L
Paul K Payne
The True Capabilities of the Nikon P1000 Part 2. The 2000 to 3000 mm focal length
2 3
Replies
46
Views
5K
fred from nashville
F
Steve Bingham
Nikon D5500 and shutter vibration
Replies
19
Views
20
GaborF
G

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top