Luminous Landscape Forum

Raw & Post Processing, Printing => Digital Image Processing => Topic started by: CynthiaM on June 16, 2008, 09:53:22 am

Title: Noise removal?
Post by: CynthiaM on June 16, 2008, 09:53:22 am
I recently returned from a trip which included a lot of visits to a lot of beautiful but poorly lit churches.  So I have a lot of images shot with a Canon 20d with high iso (1600, 3200) and of course, a lot of noise.

I'm not sure when in the workflow is the best place to attempt noise removal?  What makes sense is to turn off noise removal completely in camera raw in an attempt to maintain any edge detail.  Or should I do minimal removal in ACR or Lightroom? Once in Photoshop, should I do the noise removal, first, or all my other adjustments and noise removal last, before sharpening?

I am using Noise Ninja for the noise.
Title: Noise removal?
Post by: DarkPenguin on June 16, 2008, 10:25:37 am
If you can get away with using ACR or LR's noise removal I would just use that.  (With the 20D I found I rarely needed more than the NR available in ACR/LR at 3200.)  Otherwise turn off sharpening and NR in ACR/LR and do it immediately in CS3 with NN.
Title: Noise removal?
Post by: francois on June 16, 2008, 11:27:47 am
Cynthia,
I perform noise removal in LR or ARC only if I can get rid of it in those tools. For heavy noise removal, I do it in Photoshop using Noiseware or Noise Ninja. In this case, I don't do any capture sharpening or noise removal in LR or ACR. I do capture sharpening after noise removal.
Title: Noise removal?
Post by: Philip Weber on June 16, 2008, 12:09:35 pm
Quote: I'm not sure when in the workflow is the best place to attempt noise removal?


I'm certainly no expert but recently started doing it at the very end in CS3. I roundtrip from LR, apply capture sharpening and then, after any/all tweaks, resize, output sharpen, flatten the TIFF, THEN take care of any noise. I used to adjust noise before the very end but didn't get as good of results. From there, I head back into LR for printing.

HOWEVER, the better looking results MAY stem from the fact that I recently switched from Noise Ninja to Noiseware. They're both quality programs but for whatever reason(s), I'm doing better with the latter.

There are probably many valid reasons why this isn't the best workflow; all I can attest to is better results than before.

Phil
Title: Noise removal?
Post by: Andy M on June 16, 2008, 02:50:23 pm
I tend to use Noiseware as soon as the file hits Photoshop.

Why sharpen noise only to soften the file later?
Title: Noise removal?
Post by: Gordon Buck on June 16, 2008, 03:33:19 pm
It makes sense to me to apply noise reduction techniques as soon as possible and prior to sharpening.  However, it is my impression that the sharpening process in ACR is different from "sharpening" or unsharp mask in Photoshop.  It seems that the ACR sharpening is considered to be a sort of "capture sharpening".  Therefore, I wonder if there is an optimum blend of ACR sharpening plus 3rd part noise reduction software.

I switched from Neat Image to Noiseware a few weeks ago and find that Noiseware works better for me on my Canon G9 high ISO files.  I don't usually use the G9 at high ISO so have only limited experience with Noiseware.  However, based on a limited number of comparisons, I can't see much difference between retaining or omitting ACR sharpening when Noiseware is applied immediately thereafter.  

I plan to make some additional comparisons because it does seem illogical to sharpen the noise and then try to remove it but I don't know for certain that ACR sharpening actually sharpens the noise in a raw file.
Title: Noise removal?
Post by: Philip Weber on June 16, 2008, 07:38:14 pm
I would like to know if ACR/LR and/or PK Sharpener "sharpens noise" as well!

I didn't mention in my earlier post that I use Jeff's PhotoKit Sharpener plug-in.

As stated before, noise removal moved from before to after output sharpening only because of better looking prints but I've also found Noiseware to have a better UI (or at least one I understand better!) than Noise Ninja and so, I have been able to protect the detail better. Working with RAW files, does the process of flattening the file in CS3 after all the tweaks affect noise levels?

As I'm fairly new to digital image processing, I always appreciate the input from the more seasoned photographers out there...that's what makes the LL forum so great for someone with my experience level.

Maybe I should be hitting the noise, flattening the file and then adjusting the image size and performing the output sharpening. I end up saving as a TIFF, with LZW compression, before printing.

OR...maybe I should just say the heck with it, they look good enough to me and spend more time out taking the shots rather than staying at home tweaking them!

As always, I appreciate everyone's insights...
Phil
Title: Noise removal?
Post by: gmitchel on June 17, 2008, 08:23:27 am
I'm going to reply to a couple of threads with one reply.

(1) Sharpening does not add noise. It can emphasize any noise that's already present and take it from being insignificant to very noticeable.

(2) There is no practical way to combine noise reduction with third-party plug-ins and capture sharpening with ACR or LR. Noise reduction should occur prior to sharpening. Noise reduction is antagonistic to sharpening. Capture sharpening is gentle. You are sharpening to restore detail lost from digital capture and no more. That means noise reduction, even with a surface mask, will likely clobber capture sharpening if noise reduction is applied after capture sharpening.

(3) I have not found noise reduction in ACR or LR to be a capable solution for anything but light noise. The Canon 20D is certainly not capable of taking pictures at 1600 ISO with noise light enough that ACR or LR can effectively remove it. I find the 20D to be noisy enough even at 100 ISO to use NeatImage or Noise Ninja for noise reduction in the shadows, sky gradient, etc.

(4) If you do apply noise reduction, it is critical to use a surface mask. You want to keep noise reduction away from the edges. Otherwise you will significantly soften the photograph.

I include a chapter in both my eBook and my video on sharpening because it is so crucial to do it, do it first in the workflow, and to use a surface mask.

You can download free Photoshop tools from the Digital Darkroom of my site for adding masks to photographs, including surface masks.

http://www.thelightsright.com/digital-darkroom.htm (http://www.thelightsright.com/digital-darkroom.htm)

Cheers,

MItch
Title: Noise removal?
Post by: DarkPenguin on June 17, 2008, 08:57:29 am
Quote
(3) I have not found noise reduction in ACR or LR to be a capable solution for anything but light noise. The Canon 20D is certainly not capable of taking pictures at 1600 ISO with noise light enough that ACR or LR can effectively remove it. I find the 20D to be noisy enough even at 100 ISO to use NeatImage or Noise Ninja for noise reduction in the shadows, sky gradient, etc.

Really?  I've barely needed NR with my 20D at ISO 3200.
Title: Noise removal?
Post by: gmitchel on June 17, 2008, 09:34:18 am
Quote
Really?  I've barely needed NR with my 20D at ISO 3200.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=202053\")

Really. I only shoot ISO 100 with the 20D unless extreme need compels me to use a faster ISO.

I have owned a 20D for four years. It has less noise than most other cameras from its generation, although more than the 1D MkII, 1Ds MkII, etc. It is very noisy at ISO 3200.

Here's an independent evaluation that is the same result that I see:

[a href=\"http://www.dpreview.com/reviews/canoneos20d/page20.asp]http://www.dpreview.com/reviews/canoneos20d/page20.asp[/url]

Noise is very evident at ISO 800, ISO 1600, and ISO 3200. I have plenty of ISO 100 images in daylight with noise in shadows, sky gradients, etc. Can I remove it. Yes. I use something like NeatImage or Noise Ninja together with a surface mask.

If you cannot see noise at ISO 1600 or ISO 3200 with a Canon 20D, then I respectfully suggest you do not have a trained eye for noise. It does take experience to develop an eye for artifacts like noise. I'm not trying to insult you. I'm just objectively stating that the Canon 20D is not close to noise-free at ISO 1600 or ISO 3200. Not even the 1Ds MkII is close to noise-free at those ISOs.

It is also important to realize that most people resort to such an ISO to get enough shutter speed to handhold the camera. The longer the exposure or the longer the burst of photos, the more heat will build up in the camera and the more noise will be generated.

If you take just one isolated photo at ISO 3200 in broad daylight and use a shutter speed of 1/1000 of a second, you might have *RELATIVELY* little noise. But that's an atypical use of such ISO. The more typical use is a wedding photographer or event photographer trying to use natural lighting indoors with big bursts of photos. Guarantted, the 20D is noisy in that ciurcumstance. If you doubt it, just read any busy wedding forum re. the 20D for weddings.

Cheers,

Mitch
Title: Noise removal?
Post by: DarkPenguin on June 17, 2008, 10:45:29 am
Unless I'm dealing with deep shadows I haven't had any issue with noise that matters in a print.  Perhaps I print smaller sizes than you do.

(Comment only in regards to the Canon 20D.)
Title: Noise removal?
Post by: Misirlou on June 17, 2008, 11:31:50 am
I'm a dinosaur from the film days. I learned digital processing on raw files from my old Canon S50 P&S (which I still use, BTW). Later, I bought a 20D, and now a 40D. But I still go to 6X6 or 4X5 film when I want to shoot something for printing really large.

When I got the S50, noise reduction s/w was sort of new. I found that Neat Image made a massive improvement on S50 files, as one might expect. When I got the 20D, I just kept using the same sort of workflow. I always ran noise reduction, even at 100 ISO, as the very first step after bringing the raw file into Photoshop. I could easily see a big improvement at 1:1 on the monitor. When I got the 40D, I continued with the same workflow.

Things started to get complicated with the arrival of Lightroom. One had to decide where to fit the noise reduction in with capture sharpening. It does not help that you can't confine noise reduction to a masked area in Lightroom.

But it finally dawned on me that it might be useful to see how the noise actually effected prints. Based on my experiments (and keep in mind that I rarely print digital captures from DSLRS bigger than 8X10), noise that seems really objectionable at 1:1 on the monitor may not be meaningful in a (small) print. I think what happened is that I got so used to looking at 5MP shots at 1:1 I that kept doing that with 10MP shots out of habit, even though the 10MP shots require far less enlargement to print.

In any case, I would always do noise reduction as early as possible in the flow. Sharpening will amplify noise, and any sort of color manipulation will operate on chroma noise just as as it will on signal data. No sense in doing any of that if you can avoid it. Lately, I've been running noise reduction in Lightroom, which can handle ordinary 100 ISO noise from a 40D pretty well. Not as well or as controlably as Neat Image in Photoshop, but well enough for 8X10 prints. I'm also running DXO lately, and that does a pretty good job of noise reduction and capture sharpening in one step, although it also suffers from lack of masking.
Title: Noise removal?
Post by: gmitchel on June 17, 2008, 01:14:50 pm
Quote
But it finally dawned on me that it might be useful to see how the noise actually effected prints. Based on my experiments (and keep in mind that I rarely print digital captures from DSLRS bigger than 8X10), noise that seems really objectionable at 1:1 on the monitor may not be meaningful in a (small) print. I think what happened is that I got so used to looking at 5MP shots at 1:1 I that kept doing that with 10MP shots out of habit, even though the 10MP shots require far less enlargement to print.
[a href=\"index.php?act=findpost&pid=202089\"][{POST_SNAPBACK}][/a]

I think you'll find les need for noise reduction, especially at ISO 400 and below with the Canon 40D. It has much less noise than the 20D.

It is certainly true that print size matters. Noise is likely to be objectionable in a smaller print.

I no not edit at 100% magnification. That is very misleading for assessing noise, sharpening, etc. In the case of a 1Ds MkII or 1Ds MkIII image, you end up with a poster size image at monitor resolution. I use a zoom that approximates print size without lots of aliasing effects. Something like 50% or 25%.

Cheers,

Mitch
Title: Noise removal?
Post by: Misirlou on June 17, 2008, 01:47:34 pm
Quote
I think you'll find les need for noise reduction, especially at ISO 400 and below with the Canon 40D. It has much less noise than the 20D.

It is certainly true that print size matters. Noise is likely to be objectionable in a smaller print.

I no not edit at 100% magnification. That is very misleading for assessing noise, sharpening, etc. In the case of a 1Ds MkII or 1Ds MkIII image, you end up with a poster size image at monitor resolution. I use a zoom that approximates print size without lots of aliasing effects. Something like 50% or 25%.

Cheers,

Mitch
[a href=\"index.php?act=findpost&pid=202106\"][{POST_SNAPBACK}][/a]

Yes, I agree. I guess I'm still feeling a little burned by certain posters here who insisted a few months ago that the 40D performed no better than the 20D, despite the test shots I uploaded which looked pretty conclusive to my eye. My general experience matches yours.

In any case, to get back to the original poster's question, I feel that noise reduction should be done as early as possible in the workflow. If you can take care of it in the raw converter, great. If not, say high ISO situations, try not to sharpen very much if you can, until you deal with the noise. Just don't agonize over noise too much, because it may not be a big deal in your prints anyway.
Title: Noise removal?
Post by: bjanes on June 17, 2008, 04:31:14 pm
Quote
I no not edit at 100% magnification. That is very misleading for assessing noise, sharpening, etc. In the case of a 1Ds MkII or 1Ds MkIII image, you end up with a poster size image at monitor resolution. I use a zoom that approximates print size without lots of aliasing effects. Something like 50% or 25%.
[a href=\"index.php?act=findpost&pid=202106\"][{POST_SNAPBACK}][/a]

Judging sharpening on screen is problematic, since the screen is a low resolution device (~90 ppi) and printers have a much higher resolution (typically 300 ppi). At a screen resolution of 25 or 50%, the sharpening halos may not show up. The current version of ACR does not even allow preview of sharpening or NR below 100% and our friend Jeff Schewe suggests that you make the image sharpening look good at 100%. This applies for capture sharpening, so printer resolution is less important than for output sharpening. Still, there appears to be no consensus on judging capture sharpening on screen and comments are invited.

Bill
Title: Noise removal?
Post by: DarkPenguin on June 17, 2008, 05:01:45 pm
Quote
Judging sharpening on screen is problematic, since the screen is a low resolution device (~90 ppi) and printers have a much higher resolution (typically 300 ppi). At a screen resolution of 25 or 50%, the sharpening halos may not show up. The current version of ACR does not even allow preview of sharpening or NR below 100% and our friend Jeff Schewe suggests that you make the image sharpening look good at 100%. This applies for capture sharpening, so printer resolution is less important than for output sharpening. Still, there appears to be no consensus on judging capture sharpening on screen and comments are invited.

Bill
[a href=\"index.php?act=findpost&pid=202126\"][{POST_SNAPBACK}][/a]

I simply try to go with "no halos".  Sometimes it is "no halos that I noticed.  #$%@!#$%!!"

I'm willing to print with pretty significant halos.
Title: Noise removal?
Post by: Misirlou on June 17, 2008, 06:23:28 pm
Quote
I simply try to go with "no halos".  Sometimes it is "no halos that I noticed.  #$%@!#$%!!"

I'm willing to print with pretty significant halos.
[a href=\"index.php?act=findpost&pid=202130\"][{POST_SNAPBACK}][/a]

I thought the current theory was that you actually need some halos with inkjet printing to make up for the inevitable ink spread on the paper. That's one of the reasons you sharpen at both the capture and output stages, with entirely different goals in mind, and entirely different sharpening processes.

I'm guessing, based on hints here an elsewhere, that Adobe will be addressing output-specific sharpening in the final release of Lightroom 2.0, and we can all hope they'll be looking at a more sophisticated noise reduction approach too. And don't worry if you aren't using Lightroom, because ACR and Lightroom seem to be cross-pollinating the good ideas of late.
Title: Noise removal?
Post by: DarkPenguin on June 17, 2008, 07:08:06 pm
I think that is correct.  But you definitely do not want them at the capture sharpening stage.

Lightroom appears to have some form of pk sharpener available for output.  If I was willing to use it more I'd have more info on it.
Title: Noise removal?
Post by: Gordon Buck on June 17, 2008, 09:52:02 pm
Quote
It makes sense to me to apply noise reduction techniques as soon as possible and prior to sharpening.  However, it is my impression that the sharpening process in ACR is different from "sharpening" or unsharp mask in Photoshop.  It seems that the ACR sharpening is considered to be a sort of "capture sharpening".  Therefore, I wonder if there is an optimum blend of ACR sharpening plus 3rd part noise reduction software.

I switched from Neat Image to Noiseware a few weeks ago and find that Noiseware works better for me on my Canon G9 high ISO files.  I don't usually use the G9 at high ISO so have only limited experience with Noiseware.  However, based on a limited number of comparisons, I can't see much difference between retaining or omitting ACR sharpening when Noiseware is applied immediately thereafter. 

I plan to make some additional comparisons because it does seem illogical to sharpen the noise and then try to remove it but I don't know for certain that ACR sharpening actually sharpens the noise in a raw file.
[a href=\"index.php?act=findpost&pid=201945\"][{POST_SNAPBACK}][/a]

Replying to myself for continuity ...

I made additional comparisons using my Canon G9 raw files at ISO 400 and 800, ACR with and without sharpening (and sometimes with ACR noise reduction), PK Capture Sharpener, CS3 "Smart Sharpen" and Noiseware.  As a result of these comparisons, I'll be *capture* sharpening in ACR prior to using Noiseware.  

Fully realizing the subjective nature in the compromise between noise reduction and sharpening, I'm simply reporting what I did along with my conclusions.  I also realize that my conclusions are counter-intuitive; however, the results were very obvious.

I've had a full up-to-date copy of NeatImage for several years but have recently found Noiseware more to my liking.  In every comparison, I used Noiseware in the default mode but realize that even better results can be obtained with a little more practice and study.

I've also had and used PKSharpener in the three stage sharpening process for several years.  I believe in this product and procedure but recently have begun to substitute ACR sharpening for PK Capture Sharpen.

For all comparisons, I first made an unsharpened, no noise reduction background image in ACR/CS3.  Then I repeatedly opened the raw file with variations in ACR, processed it and pasted it as a layer over the base background.  I could make comparisons, typically at 50% size and 100% size, by turning layers on and off.  (Got to learn and get confidence in Smart Objects!)

Yes, I know that viewing prints is different from pixel peeping.  I've previously concluded that I can get acceptable -- to me -- 8x10 prints from my G9 at ISO400 with Noiseware.  

In ACR, the sharpening that I'm calling "capture" is mid-range on the Amount - typically 80%, a radius of 1, Detail from 30 to 50%, Mask from 0 to 30%.  I don't claim to be an expert or even to necessarily understand the significance of these numbers -- but I did try a lot of them!

I tried Noiseware on top of ACR Sharpening with ACR noise reduction (lum=80 and color=100) but those were obviously overdone.

I tested the maximums of ACR Sharpening followed by Noiseware.  These were obviously oversharpened but the efffect on noise reduction wasn't as bad as I thought it would be.

I tried zero sharpening and zero noise reduction in ACR followed immediately by Noiseware and then PK Capture Sharpener .  The results were OK but I preferred the ACR sharpening followed by Noiseware.  (Note:  edited 6/19/2008 because the order was transposed in the original posting.)

I tried zero sharpening and zero noise reduction in ACR followed by Noiseware with extra sharpening but didn't that that result as well as with the ACR pre-sharpening.

I tried modest ACR sharpening plus modest ACR noise reduction followed by Noiseware.  Frankly, it was pretty difficult to make a call on some of these.  

Is ACR sharpening that good?  Is Noiseware so good that it makes up for pre-sharpening?  I don't know but am hoping that others will make a few comparisons and report their findings.
Title: Noise removal?
Post by: bjanes on June 18, 2008, 12:51:01 pm
Quote
Replying to myself for continuity ...

I made additional comparisons using my Canon G9 raw files at ISO 400 and 800, ACR with and without sharpening (and sometimes with ACR noise reduction), PK Capture Sharpener, CS3 "Smart Sharpen" and Noiseware.  As a result of these comparisons, I'll be *capture* sharpening in ACR prior to using Noiseware. 
[a href=\"index.php?act=findpost&pid=202176\"][{POST_SNAPBACK}][/a]

Fine if that works for you, but your approach is contrary to all expert opinion that I have seen. You have not presented any of your results, but I wonder if your testing is sufficiently intensive to reach your conclusion.



Quote
I've also had and used PKSharpener in the three stage sharpening process for several years.  I believe in this product and procedure but recently have begun to substitute ACR sharpening for PK Capture Sharpen.

I tried zero sharpening and zero noise reduction in ACR followed by PK Capture Sharpener and then Noiseware.  The results were OK but I preferred the ACR sharpening followed by Noiseware.

I tried zero sharpening and zero noise reduction in ACR followed by Noiseware with extra sharpening but didn't that that result as well as with the ACR pre-sharpening.

I tried modest ACR sharpening plus modest ACR noise reduction followed by Noiseware.  Frankly, it was pretty difficult to make a call on some of these. 

Is ACR sharpening that good?  Is Noiseware so good that it makes up for pre-sharpening?  I don't know but am hoping that others will make a few comparisons and report their findings.
[a href=\"index.php?act=findpost&pid=202176\"][{POST_SNAPBACK}][/a]

One possibility that you did not test is to do the  capture sharpening in Noiseware. As you know, Noiseware has options for detail protection and detail enhancement. How it does these things under the hood is unclear to me--do they use masks or other algorithms? At least, I would imagine that the NR and sharpening are designed to work together.

I've done quite bit of available light shooting with the Nikon D200 at ISO 1600, where there is quite a bit of noise, even with optimum exposure and Noiseware is my favored program, although I also have NeatImage and NoiseNinja. In his sharpening book, Bruce Fraser pointed out that many noise reduction algorithms produce images that can not subsequently be sharpened without bringing back the noise in full glory. Personally, I found this to be the case with Noiseware and PK capture sharpening. For high volume work in shooting sporting events, I found the best approach was to do the NR and sharpening in Noiseware. This apporoach allows parametric editing, NR, and sharpening in Photoshop with ACR without the necessity of producing intermediate TIFF files.

Bill
Title: Noise removal?
Post by: CynthiaM on June 19, 2008, 08:34:16 am
Thanks for all of your responses.  I think I have a better handle on how to handle noise.
Title: Noise removal?
Post by: Gordon Buck on June 19, 2008, 11:54:58 am
Quote
Fine if that works for you, but your approach is contrary to all expert opinion that I have seen. You have not presented any of your results, but I wonder if your testing is sufficiently intensive to reach your conclusion.[a href=\"index.php?act=findpost&pid=202264\"][{POST_SNAPBACK}][/a]

No doubt my Photoshop skills need improving and will improve but I strongly suspect that many people making a simple comparison using light ACR sharpening followed by Noiseware will be pleasantly surprised at the results.  At the very worse, the results are OK and, I'd say, preferable.  

Note that my original post has been edited.  Of course, one of the variations was without ACR sharpening immediately followed by PK Capture Sharpener and then Noiseware.  In the original post, I transposed the order.

I've actually tested quite a few files and will be testing more.
Title: Noise removal?
Post by: Mark D Segal on June 22, 2008, 11:04:53 pm
Quote
Fine if that works for you, but your approach is contrary to all expert opinion that I have seen. You have not presented any of your results, but I wonder if your testing is sufficiently intensive to reach your conclusion.

In his sharpening book, Bruce Fraser pointed out that many noise reduction algorithms produce images that can not subsequently be sharpened without bringing back the noise in full glory. Personally, I found this to be the case with Noiseware and PK capture sharpening. For high volume work in shooting sporting events, I found the best approach was to do the NR and sharpening in Noiseware. This apporoach allows parametric editing, NR, and sharpening in Photoshop with ACR without the necessity of producing intermediate TIFF files.

Bill
[a href=\"index.php?act=findpost&pid=202264\"][{POST_SNAPBACK}][/a]

Gordon could be getting the results he talks about because the kind of sharpening he's doing in ACR does not affect smooth areas or areas with little detail, such as skies, skin and very dark tones, which is where most of the noise usually shows. It's really hard to conceive of any sharpening that sharpens noise being "a good thing" regardless of what tools one uses.

As for the combination of Noiseware and PK Capture Sharpen, this is my usual combination whenever I need to reduce noise (not often these days) and it works well. I have Noiseware set to quite high values for noise reduction e.g. 70~80% (but usually defaults for noise identification), maximum detail protection; then in PK Capture Sharpen I often use the Superfine setting eventhough the image may not look as if it has much high-frequency information. With this combination of settings the noise does not return as a result of capture sharpening.

I find it useful to examine my 1Ds3 files at both 50% and 100% display magnification, as each allows one to see different things - be it the overall likely result in a 13*19 inch print or fine artifacts as the case may be.
Title: Noise removal?
Post by: bjanes on June 23, 2008, 10:00:51 am
Quote
As for the combination of Noiseware and PK Capture Sharpen, this is my usual combination whenever I need to reduce noise (not often these days) and it works well. I have Noiseware set to quite high values for noise reduction e.g. 70~80% (but usually defaults for noise identification), maximum detail protection; then in PK Capture Sharpen I often use the Superfine setting eventhough the image may not look as if it has much high-frequency information. With this combination of settings the noise does not return as a result of capture sharpening.

I find it useful to examine my 1Ds3 files at both 50% and 100% display magnification, as each allows one to see different things - be it the overall likely result in a 13*19 inch print or fine artifacts as the case may be.
[a href=\"index.php?act=findpost&pid=202946\"][{POST_SNAPBACK}][/a]

Mark,

Or differences could be related to the camera. Your 1Ds3 has relatively low noise, whereas my Nikon D200 has high noise, especially at ISO 1600 and in the blue channel when one is shooting in tungsten illumination. Here is a representative shot taken under tungsten illimination at ISO 1600 and which was slightly underexposed and needed some positive exposure compensation in ACR. The blue channel is on the right. The RGB shows yellow splotches on the skin where there is essentially no blue signal.

[attachment=7173:attachment]

And here is a composite shot showing various NR and sharpening combinations. On the right is the default NoiseWare (sharpening = 5). The middle image shows the default NW with PK superfine sharpening. On the left, I set sharpening to zero in NW and sharpened with PK superfine.

[attachment=7174:attachment]

Note the re-appearance of noise along the hairline, ears, eyes, nose and other areas. Personally, I prefer the default NW image. Of course, there are a large number of possible permutations, and I used defaults to simply matters. All images are at 100%.

I now have a D3 and find that NR is often not needed even at ISO 3200 and my current situation may be similar to yours. Thanks for your input, which is always appreciated.

Bill
Title: Noise removal?
Post by: Mark D Segal on June 23, 2008, 10:25:00 am
Bill,

Yes indeed, both the 1Ds3 and the D3 show remarkably little noise at high ISO and that makes a huge difference to the post-processing challenge. That blue channel illustration you provide is quite a shocker; but for the majority of people who can't or won't put high megabucks into a DSLR and still expect good results at high ISO, this is approximately what they can expect to get, especially with under-exposure, so really best practice with exposure and post-capture workflow becomes all the more important.

Looking at the three extracts in your second illustration without reading what they are (to minimize "intellectual contamination") just visually on my 1600*1200 resolution LaCie 321 none of them stands out as really better than the others. You would probably see more differentiation using the original full resolution images.  But in all cases, I would assume you are doing the noise reduction before sharpening, right?

Mark
Title: Noise removal?
Post by: bjanes on June 23, 2008, 12:05:39 pm
Quote
Bill,

Yes indeed, both the 1Ds3 and the D3 show remarkably little noise at high ISO and that makes a huge difference to the post-processing challenge. That blue channel illustration you provide is quite a shocker; but for the majority of people who can't or won't put high megabucks into a DSLR and still expect good results at high ISO, this is approximately what they can expect to get, especially with under-exposure, so really best practice with exposure and post-capture workflow becomes all the more important.
[a href=\"index.php?act=findpost&pid=203063\"][{POST_SNAPBACK}][/a]

Mark,

Actually, the shot that I chose is approaching worst case, since it was underexposed and the illumination was about 2000K (for WB in ACR). With better exposure and in daylight, the blue channel is considerably better. Nonetheless, with the D200 at ISO 1600 exposure to the right is critical for best signal:noise. However, in difficult conditions, ETTR can run counter to getting enough shutter speed and one does not always get optimal exposure in action shots.

With the D3 at high ISO, I usually use the built in ACR NR if needed (usually not), followed by the built in ACR sharpening (which, as you know, is based on Bruce Fraser's workflow as is PKSharpener). I often prefer the fine grained noise pattern of the D3 to the softness introduced by NR.  The PK output sharpening works well with this workflow. Is that what you are finding with the 1Ds3?

Quote
Looking at the three extracts in your second illustration without reading what they are (to minimize "intellectual contamination") just visually on my 1600*1200 resolution LaCie 321 none of them stands out as really better than the others. You would probably see more differentiation using the original full resolution images.  But in all cases, I would assume you are doing the noise reduction before sharpening, right?

Mark
[a href=\"index.php?act=findpost&pid=203063\"][{POST_SNAPBACK}][/a]

The differences did not show as well as I had hoped, so here is another shot with NoiseWare defaults + PK fine edge on the left and NoiseWare detaults only on the right and at 200%. I uploaded to my web site as a PNG to avoid JPEG artifacts. Yes, I always do NR before any sharpening.

Bill

(http://bjanes.smugmug.com/photos/318401260_TwZWm-O.png)
Title: Noise removal?
Post by: Mark D Segal on June 23, 2008, 01:01:44 pm
Bill,

Yes, more obvious now. I do find it peculiar that a sharpening algorythm should pick up noise that has already been "reduced", but your evidence seems to point that it may. This must mean that the noise reduction is not adequate for certain edges to remain invisible to the sharpening. But then, if one reduced the noise even more (i.e. pushed well beyond the defaults) one may begin to impair desirable detail even with capture sharpening. Careful compromises!

Mark
Title: Noise removal?
Post by: Gordon Buck on June 23, 2008, 01:39:35 pm
Quote
Yes, I always do NR before any sharpening.
[a href=\"index.php?act=findpost&pid=203090\"][{POST_SNAPBACK}][/a]

Bill, realizing that ACR sharpening prior to noise reduction is not your recommendation, could you process the same image with modest ACR "capture" sharpening followed by Noiseware default?
Title: Noise removal?
Post by: gmitchel on June 24, 2008, 09:15:05 am
Quote
Bill,

Yes, more obvious now. I do find it peculiar that a sharpening algorythm should pick up noise that has already been "reduced", but your evidence seems to point that it may. This must mean that the noise reduction is not adequate for certain edges to remain invisible to the sharpening. But then, if one reduced the noise even more (i.e. pushed well beyond the defaults) one may begin to impair desirable detail even with capture sharpening. Careful compromises!

Mark
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=203110\")

I'm not surprised.

Noise reduction software doesn't remove noise. It makes it less obvious. It can be accomplished in a number of ways, but the essence is the reverse of sharpening. Neighboring pixels are examined to determine whether the values for a pixel are significantly different *WHERE* little variation is expected in terms of color and/or luminosity. The contrast between those pixels and their neighbors is reduced.

The Median filter, for example, is a fairly blunt tool for noise reduction. The neigboring pixels are consulted and their median value (sort the values and take the middle value) is used to adjust the pixel. This "smooths out" variation and reduces noise.

Sophisticated tools like NeatImage, Noise Ninja, and Noiseware have more sophisticated algorithms for determining noisy pixels and for replacing their values, but in essence they reduce contrast between noisy pixels and their neighbors.

Sharpening software can still detect the contrast between noisy pixels and their neighbors *AND* emphasize the contrast, undoing some/all of the noise reduction.

That's why I refer to noise reduction and capture sharpening as antagonistic in my eBook and my video on sharpening.

[a href=\"http://www.thelightsright.com/SharpeningYourPhotographs.html]http://www.thelightsright.com/SharpeningYourPhotographs.html[/url]

You can reduce noise to the point where capture sharpening cannot make it more visible. The resulting artifacts -- a plastic-like appearance from loss of significant details -- will likely be worse than some residual noise.

If you do not use layers masks, capture sharpening and noise reduction will largely cancel out. Depending which is more aggressive in effect, you will either end up sharpening noise or softening significant details.

Capture sharpening is best done with a layer mask that restricts sharpening to the edges. ACR and LR now include some edge masking. Focal Blade and Photokit Sharpener give you direct control over edge masking. So do my actions and scripts (there are free actions and scripts for noise reduction, masking, and sharpening on my site).

I strongly recommend noise reduction on a layer and using a surface mask to keep noise reduction away from the edges. That will allow you to use more aggressive noise reduction settings in many cases without losing significant details. That can make it less likely for noise to reemerge after capture sharpening. I have a free script that can automate the generation of surface masks. There are even options to limit the masks to tonal/color ranges (e.g., the shadows).

It takes more work, but if you can restrict noise reduction to just the features where it's most needed, like the shadows, you can be more aggressive with the noise reduction settings. Restricting noise reduction was the #1 motivation behind my masking script.

http://www.thelightsright.com/TLRProfessionalMaskToolkit.htm (http://www.thelightsright.com/TLRProfessionalMaskToolkit.htm)

Enjoy!

Mitch
Title: Noise removal?
Post by: CynthiaM on June 24, 2008, 09:42:33 am
For those of you who have not checked out Mitch's masks, do so.  They are terrific and better yet, free.
Quote
I strongly recommend noise reduction on a layer and using a surface mask to keep noise reduction away from the edges. That will allow you to use more aggressive noise reduction settings in many cases without losing significant details. That can make it less likely for noise to reemerge after capture sharpening. I have a free script that can automate the generation of surface masks. There are even options to limit the masks to tonal/color ranges (e.g., the shadows).

It takes more work, but if you can restrict noise reduction to just the features where it's most needed, like the shadows, you can be more aggressive with the noise reduction settings. Restricting noise reduction was the #1 motivation behind my masking script.

http://www.thelightsright.com/TLRProfessionalMaskToolkit.htm (http://www.thelightsright.com/TLRProfessionalMaskToolkit.htm)


[a href=\"index.php?act=findpost&pid=203335\"][{POST_SNAPBACK}][/a]
Title: Noise removal?
Post by: Mark D Segal on June 24, 2008, 11:47:11 am
Quote
I'm not surprised.

Noise reduction software doesn't remove noise. It makes it less obvious. It can be accomplished in a number of ways, but the essence is the reverse of sharpening. Neighboring pixels are examined to determine whether the values for a pixel are significantly different *WHERE* little variation is expected in terms of color and/or luminosity. The contrast between those pixels and their neighbors is reduced.

The Median filter, for example, is a fairly blunt tool for noise reduction. The neigboring pixels are consulted and their median value (sort the values and take the middle value) is used to adjust the pixel. This "smooths out" variation and reduces noise.

Sophisticated tools like NeatImage, Noise Ninja, and Noiseware have more sophisticated algorithms for determining noisy pixels and for replacing their values, but in essence they reduce contrast between noisy pixels and their neighbors.

Sharpening software can still detect the contrast between noisy pixels and their neighbors *AND* emphasize the contrast, undoing some/all of the noise reduction.

That's why I refer to noise reduction and capture sharpening as antagonistic in my eBook and my video on sharpening.

http://www.thelightsright.com/SharpeningYourPhotographs.html (http://www.thelightsright.com/SharpeningYourPhotographs.html)

You can reduce noise to the point where capture sharpening cannot make it more visible. The resulting artifacts -- a plastic-like appearance from loss of significant details -- will likely be worse than some residual noise.

If you do not use layers masks, capture sharpening and noise reduction will largely cancel out. Depending which is more aggressive in effect, you will either end up sharpening noise or softening significant details.

Capture sharpening is best done with a layer mask that restricts sharpening to the edges. ACR and LR now include some edge masking. Focal Blade and Photokit Sharpener give you direct control over edge masking. So do my actions and scripts (there are free actions and scripts for noise reduction, masking, and sharpening on my site).

I strongly recommend noise reduction on a layer and using a surface mask to keep noise reduction away from the edges. That will allow you to use more aggressive noise reduction settings in many cases without losing significant details. That can make it less likely for noise to reemerge after capture sharpening. I have a free script that can automate the generation of surface masks. There are even options to limit the masks to tonal/color ranges (e.g., the shadows).

It takes more work, but if you can restrict noise reduction to just the features where it's most needed, like the shadows, you can be more aggressive with the noise reduction settings. Restricting noise reduction was the #1 motivation behind my masking script.

http://www.thelightsright.com/TLRProfessionalMaskToolkit.htm (http://www.thelightsright.com/TLRProfessionalMaskToolkit.htm)

Enjoy!

Mitch
[a href=\"index.php?act=findpost&pid=203335\"][{POST_SNAPBACK}][/a]

Hi Mitch,

Yes, this is all very much to the point. I always use PK Sharpener, and most of the time I confine noise reduction to the areas of an image where it shows most by using a layer and layer mask. One then also has the added freedom of playing with opacities to get the right blend between sharpening and noise reduction in the event of areas where some compromise is inevitable. More often than not, given where noise usually shows most, it is not. Thanks for the reference to your toolkit.

Mark
Title: Noise removal?
Post by: bjanes on June 24, 2008, 06:02:46 pm
Quote
Bill, realizing that ACR sharpening prior to noise reduction is not your recommendation, could you process the same image with modest ACR "capture" sharpening followed by Noiseware default?
[a href=\"index.php?act=findpost&pid=203123\"][{POST_SNAPBACK}][/a]


This is what you requested on the left. The image with NR in Noiseware and then capture sharpening in PK (the same image as previously shown) is on the right. IMHO, the ACR sharpened image followed by Noiseware has less edge noise, possibly because of differences in masking or sharpening settings, than other image.

Optimum results would probably involve NR with Noiseware using a layer and surface mask followed by sharpening on a layer with an edge mask.

Bill

(http://bjanes.smugmug.com/photos/319305217_vo2Wv-O-3.png)
Title: Noise removal?
Post by: gmitchel on June 24, 2008, 06:51:24 pm
Quote
Thanks for the reference to your toolkit.
[a href=\"index.php?act=findpost&pid=203372\"][{POST_SNAPBACK}][/a]

I hope the mask toolkit helps. I always use a surface mask for noise reduction. Then I can grab a brush filled with black at something like 20% opacity and further adjust the mask so that noise reduction is applied where it's most needed. Shadows or gradients for sky or water, mostly.

The surface mask is not just an inverted edge mask, BTW. That could make for some artifacting. The surface mask has a bit more Gaussian Blur applied.

Cheers,

Mitch
Title: Noise removal?
Post by: Gordon Buck on June 24, 2008, 09:40:30 pm
Quote
This is what you requested on the left. ... [a href=\"index.php?act=findpost&pid=203455\"][{POST_SNAPBACK}][/a]

Thanks, Bill.  You have a finer and more critical eye than I.  I'd say those two treatments produce essentially the same result.  That being the case, I'd stay with the ACR sharpening followed by Noiseware.  But I suspect you are correct about the optimum; I'm certainly still seeking that optimum.

How do you work with Noiseware for distant detail like leaves in trees?  I see that there is a landscape setting and note the differences in the various sliders.  It seems to me that noise reduction for distant detail is likely to be a different layer with very light settings, especially for luminance.
Title: Noise removal?
Post by: Mark D Segal on June 24, 2008, 11:38:27 pm
Quote
This is what you requested on the left. The image with NR in Noiseware and then capture sharpening in PK (the same image as previously shown) is on the right. IMHO, the ACR sharpened image followed by Noiseware has less edge noise, possibly because of differences in masking or sharpening settings, than other image.

Optimum results would probably involve NR with Noiseware using a layer and surface mask followed by sharpening on a layer with an edge mask.

Bill

(http://bjanes.smugmug.com/photos/319305217_vo2Wv-O-3.png)
[a href=\"index.php?act=findpost&pid=203455\"][{POST_SNAPBACK}][/a]

Bill,

from what I see, both images are artifacted, but in different ways, and I'd be really hard-pressed to say which I prefer, because they're both kind of.......OK you get the point.

Over the past few days I did a bunch of grandchildren pics - shot them at ISO 1600 with the 1Ds3 because I wanted to further explore the question of noise on skin at high ISO with reasonably correct exposures. The camera really takes specular highlights seriously so I had to force it to be "wrong" with at least 1 stop of positive EC, which took me into a reasonable range of ETTR-land (still needed about half a stop of Exposure adjustment in ACR, and some Brightness). At 50% magnification on display the noise is really slight, but shows more at 100%. It's  a fine-grain effect. Noiseware shows it as low level high-frequency noise ranging from about 5 in the highlights to 10~12 in the shadows. I ran default Noiseware on the skin only, avoiding eyes and lips. Then I ran default PK Capture for High Res Digital at Superfine. It really didn't bring the noise back. With all this stuff on layers I can switch anything on and off in any sequence, which I've done, and I've observed that the combination of NW and PK works well. So this story is obviously very specific to the kind and amount of noise one is dealing with, hence to an extent camera-specific.
Title: Noise removal?
Post by: Panopeeper on June 24, 2008, 11:59:03 pm
Bill,

questions regarding the sample you posted above:

1. do the yellow blotches appear in Nikon Capture as well?

2. which step/action gets rid of them in ACR?
Title: Noise removal?
Post by: bjanes on June 25, 2008, 08:59:49 am
Quote
Bill,

questions regarding the sample you posted above:

1. do the yellow blotches appear in Nikon Capture as well?

2. which step/action gets rid of them in ACR?
[a href=\"index.php?act=findpost&pid=203495\"][{POST_SNAPBACK}][/a]

Gabor,

Yes, the yellow blotches do appear in Nikon Capture conversions, but to a lesser extent. These blotches appear where there are dark areas in the blue channel, which makes sense since yellow is the complement of blue.

With all sharpening and noise reduction turned off in both programs, the noise pattern between the two programs is quite different. The grain pattern in NC is tighter, more prominent, and uniform. In ACR setting the default color NR is 25, and this results in a more blotchy appearance in the blue channel and increased yellow blotches. Setting color NR to zero in ACR makes the pattern more uniform, but the high frequency noise is markedly increased.

For illustration, Nikon Capture NX is on the left and ACR with color NR= 0 and luminance NR = 0 is on the right:

[attachment=7211:attachment]


This view demonstrates the blue channel of ACR with ACR color NR = 0, luminance NR = 0 on the left and ACR with the default color NR = 25 on the right.

[attachment=7212:attachment]

Finally, here are the Noiseware profiles. NC is on the left (ignore CR 25 in the title), and ACR with color NR = 25 in the middle and ACR with color NR = 0 on the right.

[attachment=7213:attachment]

From this limited testing, Nikon Capture NX has better noise characteristics with the test image and the results with Noiseware NR are better with the NC image (not illustrated).
Title: Noise removal?
Post by: Mark D Segal on June 25, 2008, 09:17:23 am
Bill,

Thanks for taking the trouble to do this. Regarding the top row of illustrations, I guess it depends on whether you prefer Pointillisme or Fauvisme as a style of photographic art.  

For the second row - how do you capture a blue channel in ACR, which has no channels palette - did you get at it with Rawnalyze or another piece of software?

The Noiseware profiles are on the whole pretty ugly, with a couple of exceptions. I think all this raises a real question about whether beyond a certain level of noise -  any of these tools can really do a good job of rescuing the image while preserving wanted detail.
Title: Noise removal?
Post by: bjanes on June 25, 2008, 09:44:30 am
Quote
Bill,

Thanks for taking the trouble to do this. Regarding the top row of illustrations, I guess it depends on whether you prefer Pointillisme or Fauvisme as a style of photographic art.   

For the second row - how do you capture a blue channel in ACR, which has no channels palette - did you get at it with Rawnalyze or another piece of software?

The Noiseware profiles are on the whole pretty ugly, with a couple of exceptions. I think all this raises a real question about whether beyond a certain level of noise -  any of these tools can really do a good job of rescuing the image while preserving wanted detail.
[a href=\"index.php?act=findpost&pid=203562\"][{POST_SNAPBACK}][/a]

Mark,

For the blue channel I rendered the files with ACR into Photoshop, and got the blue channel from there.

I think you are correct in stating that the noise in this image is too high to be rescued by any type of NR. The exposure correction was +1.75 EV for this image. With proper ETTR at ISO 1600, the results are much better with the D200, but this camera has too much noise for good high ISO shooting.

Bill
Title: Noise removal?
Post by: Panopeeper on June 25, 2008, 10:47:58 am
Quote
Yes, the yellow blotches do appear in Nikon Capture conversions, but to a lesser extent. These blotches appear where there are dark areas in the blue channel, which makes sense since yellow is the complement of blue.
Bill,

do you mind uploading the raw of this demo?

Thanks
Title: Noise removal?
Post by: bjanes on June 25, 2008, 11:05:49 am
Quote
Bill,

do you mind uploading the raw of this demo?

Thanks
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=203578\")

Not at all. Please let us know about your findings.

Bill


[a href=\"https://download.yousendit.com/2B85914E523CA118]https://download.yousendit.com/2B85914E523CA118[/url]
Title: Noise removal?
Post by: gmitchel on June 25, 2008, 05:06:26 pm
I am troubled by the sample images here.

I understand the use of greatly magnified examples to distinctly point out noise to others.

However, if you are going to compare noise reduction workflows or the results of differnet noise reduction tools, the examples should be at print size (or display size for Web images).

Grossly magnified samples of noise reduction are just as misleading as grossly magnified samples of sharpening.

I have many prints where you would need a loupe to see the noise, yet at 100% on the monitor, noise is very obvious. Especially with images from cameras like the Canon 1Ds MkII, 1Ds MkIII, etc.

If the goal is to understand how noise reduction and capture sharpening interact, the magnified samples on this thread are really not all that helpful.

Cheers,

Mitch
Title: Noise removal?
Post by: Mark D Segal on June 25, 2008, 05:37:26 pm
Quote
I am troubled by the sample images here.

IIf the goal is to understand how noise reduction and capture sharpening interact, the magnified samples on this thread are really not all that helpful.

Cheers,

Mitch
[a href=\"index.php?act=findpost&pid=203671\"][{POST_SNAPBACK}][/a]

This I don't agree with. I've found that I need 100% magnification to really see what's going on with my 1Ds3 files. That doesn't mean the display view is what will show in a letter size, or even 13*19 inch print - but that's not the point. Who knows, one day I may wish to repurpose a file for yet much larger sizes, and then what's really there could show in a print as it does magnified on the display. Also, for analytical purposes it is necessary to magnify the image to the extent needed to see what the file consists of. The only risk here is confusing pixellation with sharpening artifacts and noise, but I assume those of us doing this work are aware of that and know how to tell the difference.
Title: Noise removal?
Post by: bjanes on June 25, 2008, 06:33:25 pm
Quote
I am troubled by the sample images here.

I understand the use of greatly magnified examples to distinctly point out noise to others.

However, if you are going to compare noise reduction workflows or the results of differnet noise reduction tools, the examples should be at print size (or display size for Web images).

Grossly magnified samples of noise reduction are just as misleading as grossly magnified samples of sharpening.

I have many prints where you would need a loupe to see the noise, yet at 100% on the monitor, noise is very obvious. Especially with images from cameras like the Canon 1Ds MkII, 1Ds MkIII, etc.

If the goal is to understand how noise reduction and capture sharpening interact, the magnified samples on this thread are really not all that helpful.

Cheers,

Mitch
[a href=\"index.php?act=findpost&pid=203671\"][{POST_SNAPBACK}][/a]


If I had a 360 ppi monitor, then previewing on screen at the actual print size would be a good idea. As is, I need a higher magnification.

Bill
Title: Noise removal?
Post by: gmitchel on June 25, 2008, 09:24:27 pm
Quote
This I don't agree with. I've found that I need 100% magnification to really see what's going on with my 1Ds3 files. That doesn't mean the display view is what will show in a letter size, or even 13*19 inch print - but that's not the point. Who knows, one day I may wish to repurpose a file for yet much larger sizes, and then what's really there could show in a print as it does magnified on the display. Also, for analytical purposes it is necessary to magnify the image to the extent needed to see what the file consists of. The only risk here is confusing pixellation with sharpening artifacts and noise, but I assume those of us doing this work are aware of that and know how to tell the difference.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=203678\")

Noise reduction is antagonistic towards sharpening and image details. We should, therefore, only reduce noise to the point that it is no longer a distracting visual artifact.

As George Bowen at Imagenomic told me today, "Pixel peeping at anything other than intended output size is a waste of time, especially in the noise reduction industry." Here's a .PDF from Noiseware on workflow:

[a href=\"http://www.imagenomic.com/downloads/NWNRWorkflow.pdf]http://www.imagenomic.com/downloads/NWNRWorkflow.pdf[/url]

As the PDF indicates, Print Size and 100% Magnification give two very different perceptions about noise. "You can drive yourself crazy trying to remove all noise from an image at higher screen magnifications, when you will achieve much better results at a view that corresponds to your output." [Pgs. 7-8]

George and I were chatting via e-mail because I am going to extend my reviews of noise reduction software and sharpening software.

There is no "one setting fits all resampling" setting for noise reduction software. You will almost certainly reduce noise too much by using higher magnification as your guide and obliterate fine details that could be preserved. "Be careful to not fall into the rut that some advocate of doing noise reduction at high magnification and then touting that the downward resized image will look even better. In all but minimal cases, the smaller image will lose a lot of detail using this method." [Pg. 8]

As with sharpening, your monitor is not a good guide for final sharpening settings. Prints can show noise that a monitor cannot reproduce. You need a hard proof.

You should always apply noise reduction on a layer because you might need to adjust the settings later. Tonal adjustments will affect the visibility of luminosity noise. Color adjustments will affect the visibility of color noise. You can find that your noise reduction settings were not optimal after you finish adjustments to color, tone, and sharpening. It helps to be able to go back and substitute other settings.

I'll close with this: "The key to attaining an optimum level of noise reduction is determined by your output type and size. There is no generic "one size fits all" approach that will allow optimum reduction for all of the various outputs that are available to an image. A small web viewed image will not need the same level of noise reduction as an 8x10 print, and the 8x10 print will require different reduction methods than will a 20x30 print. Furthermore, print output type, whether continuous tone or inkjet based, and paper types will play important roles." [Pg. 6]

The approach you propose is more efficient. Run noise reduction once to make a master file and then use that to resample into several different output sizes. As the "Noise Reduction Workflow Tutorial" makes very clear, efficient is not optimal in this case.

I invite you to ask Kent Christiansen at PictureCode (Noise Ninja) or Vlad at NeatImage. They'll likely give you the same advice.

Cheers,

Mitch
Title: Noise removal?
Post by: Gordon Buck on June 25, 2008, 09:53:19 pm
When my digital camera had only  3/4 MP, I looked at the images "full screen".  Is it permissible to crop my many megapixels such that the remaining cropped image is full screen?
Title: Noise removal?
Post by: gmitchel on June 25, 2008, 10:30:05 pm
Quote
When my digital camera had only  3/4 MP, I looked at the images "full screen".  Is it permissible to crop my many megapixels such that the remaining cropped image is full screen?
[a href=\"index.php?act=findpost&pid=203712\"][{POST_SNAPBACK}][/a]

At 1024x768 screen resolution, your 3/4 MP camera image was 1:1 in terms of screen pixels and image pixels. But your print resolution was still probably something like 180 ppi, which would still give you approximately a 200% zoom relative to print size.

"Learn to view your noise reduction in a manner that coincides with your intended output. Trying to remove all noise in 400% zooms will cause you to age prematurely." [Pg. 74] You were using a 200% zoom, not quite so bad. The added stress will just shorten your lifespan. LOL.

I have used a 200% zoom or even a 400% zoom to get an idea about whether noise is likely to become visible after corrections to color and tone and after sharpening. With sharpening, I pixel peep, because I can translate optimal sharpening into sharpening halo contour width. I do not use those zooms to determine my noise reduction settings.

Have you ever seen a RAW file from a Canon 1Ds MkII or MkIII at 100% zoom? It is like looking at a poster size print with your face a few inches away. Go to 200% or 400% and yikes, it's like looking at a billboard a few feet away! Tiny imperfections that the eye would almost cerftainly miss at print size stand out, and the inclination is to reduce those artifacts. That leads to smearing and loss of detail from too much noise reduction.

Using too much magnification is guaranteed to scare you, even with a camera like the 1Ds MkII or MkIII. You can see individual pixel-level abnormalities on the monitor that would require a loupe to see on a print and invisible to the eye at normal viewing distance.

Cheers,

Mitch
Title: Noise removal?
Post by: Mark D Segal on June 25, 2008, 10:58:56 pm
Quote
As George Bowen at Imagenomic told me today, "Pixel peeping at anything other than intended output size is a waste of time, especially in the noise reduction industry." Here's a .PDF from Noiseware on workflow:


Cheers,

Mitch
[a href=\"index.php?act=findpost&pid=203705\"][{POST_SNAPBACK}][/a]

Self-serving advice, and the rest of his incantation ignores the fact that people re-purpose images. He may be good at writing software, but I'm not impressed with this approach to using it. At 100% I can see both noise and image detail, especially with everything I do on separate layers, so I don't have any problem assessing the trade-off between noise reduction and image detail. The main thing in Photoshop is to use screen magnifications of 25%, 50%, 75% or 100% because the image is aliased at intermediate values.

I should add to this, that the Noiseware preview shows the image at 100% by default.
Title: Noise removal?
Post by: gmitchel on June 26, 2008, 12:15:53 am
Quote
Self-serving advice, and the rest of his incantation ignores the fact that people re-purpose images. He may be good at writing software, but I'm not impressed with this approach to using it. At 100% I can see both noise and image detail, especially with everything I do on separate layers, so I don't have any problem assessing the trade-off between noise reduction and image detail. The main thing in Photoshop is to use screen magnifications of 25%, 50%, 75% or 100% because the image is aliased at intermediate values.

I should add to this, that the Noiseware preview shows the image at 100% by default.
[a href=\"index.php?act=findpost&pid=203723\"][{POST_SNAPBACK}][/a]

LOL. Since Noiseware has no idea what size your image will be printed and at what resolution, how could it display the image at a magnification close to print size?!

Why would you gratuitously insult George by referring to his comment as self-serving?! What evidence do you have for such a gratuitous and outrageous personal attack? I don't know that George wrote the Imagenomics PDF file. It says it is authored by the Imagenomics Team.

Who wrote the PDF is irrelevant. Stop the ad hominem and deal with the logic. That same tutorial could have been written for Noise Ninja, NeatImage, etc.

If you want to claim superior judgment to the people who engineer noise reduction software, share your credentials. I doubt you have tested as many images, from as many cameras and scanners, under as many conditions as any of the people who engineer noise reduction software. Maybe you have?!

You have not adequately addressed the fact that noise is over-emphasized with magnified zooms. It is much more apparent than it is at print size. Do you deny this?!

If not, then how can reducing an over-emphasized representation of the visual impact of the noise lead to optimal noise reduction?!

You seem to assume that the relationship between noise and image details is constant at all output sizes and output media. That simply is not the case at all. You can easily prove this to yourself by taking any noisy photo and reducing it or enlarging it.

Since it is not the case that the relationship between visible noise artifacts and image detail is fixed, you can only get to optimal noise reduction for a print (which is just enough to make the noise non-apparent to the eye under expected viewing conditions and no more) by hard proofing. You can come close with a monitor by using a zoom that is near to the intended output size (near being a reference to aliasing).

If your zoom overemphasizes noise and you react to that, do you then pull back on the settings to compensate for overemphasis from your magnified view to avoid too much noise reduction?!?! I'll bet you do not. And that, Mark, means you almost certainly lose detail that could be preserved with better technique.

Cheers,

Mitch
Title: Noise removal?
Post by: Mark D Segal on June 26, 2008, 08:05:02 am
Quote
LOL. Since Noiseware has no idea what size your image will be printed and at what resolution, how could it display the image at a magnification close to print size?!

Why would you gratuitously insult George by referring to his comment as self-serving?! What evidence do you have for such a gratuitous and outrageous personal attack? I don't know that George wrote the Imagenomics PDF file. It says it is authored by the Imagenomics Team.

Who wrote the PDF is irrelevant. Stop the ad hominem and deal with the logic. That same tutorial could have been written for Noise Ninja, NeatImage, etc.

If you want to claim superior judgment to the people who engineer noise reduction software, share your credentials. I doubt you have tested as many images, from as many cameras and scanners, under as many conditions as any of the people who engineer noise reduction software. Maybe you have?!

You have not adequately addressed the fact that noise is over-emphasized with magnified zooms. It is much more apparent than it is at print size. Do you deny this?!

If not, then how can reducing an over-emphasized representation of the visual impact of the noise lead to optimal noise reduction?!

You seem to assume that the relationship between noise and image details is constant at all output sizes and output media. That simply is not the case at all. You can easily prove this to yourself by taking any noisy photo and reducing it or enlarging it.

Since it is not the case that the relationship between visible noise artifacts and image detail is fixed, you can only get to optimal noise reduction for a print (which is just enough to make the noise non-apparent to the eye under expected viewing conditions and no more) by hard proofing. You can come close with a monitor by using a zoom that is near to the intended output size (near being a reference to aliasing).

If your zoom overemphasizes noise and you react to that, do you then pull back on the settings to compensate for overemphasis from your magnified view to avoid too much noise reduction?!?! I'll bet you do not. And that, Mark, means you almost certainly lose detail that could be preserved with better technique.

Cheers,

Mitch
[a href=\"index.php?act=findpost&pid=203733\"][{POST_SNAPBACK}][/a]

Calm down. I'm not personally attacking anyone. I'm criticizing a workflow recommendation as favoring the apparent quality of the product versus that of the image.

Noiseware is an excellent product and I have every respect for it's author's software engineering skills, as I said, but when someone tells me only to look for artifactss at a magnification where they don't appear, but they may still exist, what am I supposed to take away from that, when I really don't think it's good technical advice? You still haven't dealt with my point that people re-purpose images. Once you've flattened your noise reduction layer, saved and quit, you've cooked your goose.

The logic I'm relating here is very simple - and very logical - zoom the image to the magnification needed to see what's really there. If you read what I said, I did not say that Noiseware uses a magnification reflecting print size. I said its default magnification is 100% - in the usual meaning of display magnification - i.e. display pixels, not print.

If you read what I said, you'll see that I have already acknowledged that 100% magnification will exaggerate the extent of VISIBLE noise relative to letter size and perhaps even 13*19 inch prints I would obtain from my 1Ds3, but that's not the central issue I'm getting at. I'm talking about the magnification needed to see what's really embedded in the image regardless of print size. And yes, I agree, what's VISIBLE depands on size, but that's not the point.  

I know my display, the software and print conditions well enough that I don't need to hardproof to see what I'm getting. I can optimise between noise and sharpening at 50% or 100%, then return to 25%~50% depending on print size and that gives me a reliable indicator of what VISIBLE EFFECT to expect out of the printer. This comes with some experience and lack of image detail hasn't been a complaint from professional peer reviewers.

You are making all this seem more ephemeral and difficult than it really is. Furthermore, with appropriate Noiseware settings, selective noise reduction, separate layers and the use of opacity controls, there's no reason to overcompensate either noise reduction or sharpening whether at 100% magnification or lower.

Take it easy Mitch. This little tempest in a teapot started in support of an analytical approach presented by Bill Janes which I thought quite useful and you didn't. So be it and AMEN.
Title: Noise removal?
Post by: bjanes on June 26, 2008, 09:49:27 am
Quote
Noise reduction is antagonistic towards sharpening and image details. We should, therefore, only reduce noise to the point that it is no longer a distracting visual artifact.

As with sharpening, your monitor is not a good guide for final sharpening settings. Prints can show noise that a monitor cannot reproduce. You need a hard proof.
[a href=\"index.php?act=findpost&pid=203705\"][{POST_SNAPBACK}][/a]

Your points about pixel peeping are well taken. However, the high magnification views that I posted were not to preview the appearance of noise at a given print size (in which case your criticism is valid), but rather to demonstrate what degree of noise is present in the image.


Quote
I invite you to ask Kent Christiansen at PictureCode (Noise Ninja) or Vlad at NeatImage. They'll likely give you the same advice.

[a href=\"index.php?act=findpost&pid=203705\"][{POST_SNAPBACK}][/a]

Jim Christian is the force behind Noise Ninja
Title: Noise removal?
Post by: gmitchel on June 26, 2008, 11:44:16 am
> Self-serving advice, and the rest of his incantation

That is not exactly dispassionate, Mark. That's very charged language. You could have disagreed about the workflow recommendation without the gratuitous attack. Or, perhaps you think "self-serving" and linking someone to witchcraft or alchemy is benign or neutral language?

I am and remain perfectly calm. I can disagree with you without getting animated. LOL.

> talking about the magnification needed to see what's really embedded in the
> image regardless of print size. And yes, I agree, what's VISIBLE depands on
> size, but that's not the point.

Who cares about noise embedded in the image regardless of print size?!?! What's visible is entirely the point.

> when someone tells me only to look for artifactss at a magnification where they
> don't appear, but they may still exist, what am I supposed to take away from
> that

That you need to use some common sense. As a digital photographer, what matters is the visible effect of noise on our photographs. If your eye is not distracted by the noise that's present in a photograph, where's the problem?! Leave it alone.

What does it matter, if you intend to print at 8x12 and the print has no visible noise *BUT* a 400% magnification with a loupe shows evidence of noise? How does that diminsh the print in any way. Do you print with the intended viewing distance being the eyepiece end of a loupe?! If so, then go for the extreme magnification.

Noise reduction comes at a high cost. You only want to apply enough to make noise non-apparent (or less apparent in the case of extreme noise). Any more than that and you sacrifice image detail.

What you want to do is pretend that there is one setting for noise reduction that is suitable for a master file that can be reproduced at different sizes and resolutions. That's simply not the case. Ignoring that and targeting the maximum size you might ever intend to print or trying to remove every vestige of noise will certainly erode detail for smaller prints that could have been retained by targeting the noise reduction for the intended output.


> I know my display, the software and print conditions well enough that I don't
> need to hardproof to see what I'm getting.

LOL. You had the nerve to apply "self-serving" to the comments of another!?

Rubbish. Monitors emit light. Prints reflect it. Not all noise visible on a print is visible on a monitor and vice versa. You can come close with a soft proof.


> However, the high magnification views that I posted were not to preview the
> appearance of noise at a given print size (in which case your criticism is valid),
> but rather to demonstrate what degree of noise is present in the image.

I understand all of that. You have one perspective. Mark, however, is advocating a different workflow. One that assumes a constant relationship between the appearance of noise and detail.

The comparison of how much noise exists at high magnification is still misleading, however. Since that is not the size of the intended output (maybe not even the intended resolution and media), differences in noise might become much less significant (or even insignificant) at more appropriate magnification.

It doesn't matter if the comparison results in significant differences in noise at high magnification, unless that high magnification is what you intend for the photo. It only matters if significant differences exist in the final output.

If you want to compare what happens to a photo as a result of noise reduction and capture sharpening, hold everything else constant through output and then view that output. If a Web image is the final output, look at that. You don't need magnification to see visible artifacts from noise that catches the eye. If it's not apparent to the eye, it's not worthy of comparison.

What you're doing is like the audiophile who sees noise on an oscilloscope. Who cares if you can see audio noise as a waveform?! Can you hear it, is the relevant question when you're buying audio gear.

If you cannot see a significant difference in your output, who cares if you can see it with a loupe (or the same thing, high magnification on a monitor)? Just enjoy your print and don't sweat what cannot be seen under proper viewing conditions.

Cheers,

MItch
Title: Noise removal?
Post by: gmitchel on June 26, 2008, 11:49:06 am
Quote
Jim Christian is the force behind Noise Ninja
[a href=\"index.php?act=findpost&pid=203791\"][{POST_SNAPBACK}][/a]

LOL. Yes. I've spent too much time on DPR lately. Kent Christiansen is a very active member there.

My apology to Jim.

Thanks for the nudge.

Cheers,

Mitch
Title: Noise removal?
Post by: Mark D Segal on June 26, 2008, 02:23:39 pm
Quote
> Self-serving advice, and the rest of his incantation

That is not exactly dispassionate, Mark. That's very charged language. You could have disagreed about the workflow recommendation without the gratuitous attack. Or, perhaps you think "self-serving" and linking someone to witchcraft or alchemy is benign or neutral language?

I am and remain perfectly calm. I can disagree with you without getting animated. LOL.


MItch
[a href=\"index.php?act=findpost&pid=203807\"][{POST_SNAPBACK}][/a]

No personal attack against anyone was intended; if it came accross that way I'm sorry about that; time to move on.

Turning to the substance, I think all relevant angles on this subject have been aired, and readers can test for themselves what workflow they prefer. I have nothing more to add to it.