Ad
Ad
Ad
Pages: « 1 [2] 3 »   Bottom of Page
Print
Author Topic: How Sharp Is Your Printer? How Sharp Are Your Eyes?  (Read 11220 times)
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2911


« Reply #20 on: January 07, 2010, 06:45:04 AM »
ReplyReply

Quote from: NikoJorj
As far as I understand it, the question is more adressed in the comments than in the article ; basically, Ctein just says that in mixed photographic subjects (that's what his experiment is about using a 'contact print'), aliasing is far from significant, which sound rather reasonable to me (unless of course artefacts are amplified by strong sharpening? I do also regret that output sharpening is not adressed at all in the article).
If you only shoot architectural subjets (brick walls and tiled roofs), your mileage may implode.

Scanned film grain or what is more usual the aliased version of it in a scan. If that is what you start from it is easy to create more aliasing en route. The samples that illustrate the article did ring that bell though I didn't check it on that issue.

I  have seen aliasing in more than architecture. Vegetation like grass, cane etc. Fabrics, textures.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/



Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1433


WWW
« Reply #21 on: January 07, 2010, 07:01:32 AM »
ReplyReply

Quote from: Ionaca
Co-incidentally I have been running some A4 print tests at 360 dpi on Canson Photographique 310 gsm to compare speed and print quality between three printers. Using a loupe the best dithering and dot pattern to my eyes seems to be the Epson 7900 at 2880 dpi (but only just) while the worst dithering and dot pattern to my eyes is the Epson 9600 at 720 dpi. However, without the use of a loupe I can't really tell the difference when comparing dithering and dot patterns.


The 7900 should be the best. They really worked on the masking both for the driver side and the ASIC taking advantage of 16 bit workflow too.

The substrate is of course a big factor in the acutance of the drops, hence sharpness is largely detrmined here as well.

Yet the topic seems to be going between visual appearance of sharpness distance being relative and print line shaprness at pixel peeping distances. Motifs and patterns caused by dithering are usually apparent only within very close ranges. Yet as you said on A4s this is a distance at which random patterns , motifs, etc will be within the distance to still see these artefacts.
Logged
Ryan Grayley
Full Member
***
Offline Offline

Posts: 217



WWW
« Reply #22 on: January 07, 2010, 10:26:52 AM »
ReplyReply

Quote from: neil snape
The 7900 should be the best. They really worked on the masking both for the driver side and the ASIC taking advantage of 16 bit workflow too.

The substrate is of course a big factor in the acutance of the drops, hence sharpness is largely detrmined here as well.

Yet the topic seems to be going between visual appearance of sharpness distance being relative and print line shaprness at pixel peeping distances. Motifs and patterns caused by dithering are usually apparent only within very close ranges. Yet as you said on A4s this is a distance at which random patterns , motifs, etc will be within the distance to still see these artefacts.

Sorry, I forgot to mention that the third printer I compared was the HP Z3200.
« Last Edit: January 07, 2010, 11:00:43 AM by Ionaca » Logged

Ryan Grayley BA IEng MIET ARPS
RGB Arts Ltd, London, UK
bjanes
Sr. Member
****
Offline Offline

Posts: 2882



« Reply #23 on: January 07, 2010, 02:01:20 PM »
ReplyReply

Quote from: Schewe
Well, the viewing distance doesn't actually have anything to do with the optimal output sharpening...the only two critical factors are pixel density (PPI) and media. Viewing distance does have an indirect impact in that you choose the output resolution based on your desired output size. So, bigger printed image size would equal lower PPI which would require the sharpening for the lower pixel density.

Quote from: BartvanderWolf
Hi Jeff,
I beg to differ on that. The human visual system does have a different contrast sensitivity at different spatial frequencies, and those depend on viewing distance.

One could quibble on how important it is, but stating that it has nothing to do with it is just not true.

Try this if you still don't agree ...

Kind regards,
Bart
Bart,

I'm glad to see you on LuLa. Your ImagingTechnology site is excellent, but unfortunately the forum is not that active.

Jeff's own book states (page 84), "...the secret is to keep the size of the [sharpening] halos below the threshold of visual acuity at the intended viewing distance--this is where the size of the pixels on the output becomes a critical factor"

He then goes on to say that for smaller reproductions such as used in his book (no larger than 5 x 7 inches), he tries to keep the sharpening halos to 0.01 inch, whereas for larger reproductions one can go up to 0.02 inch. If one were viewing a larger print at the same distance as for the smaller one, then it would seem as one should use the same sharpening parameters as for the smaller print. One could cut away the peripheral portion of the larger print, as it would be outside the eye's field of sharp vision in any event.

The link to Norman Koren's site regarding spatial contrast sensitivity is most appropriate for further discussion. Although the eye can resolve 60 lines per degree of arc, the frequencies around 6 cycles per degree contribute most to perceived image quality and this is the basis of the SQF measurements that Mr. Koren comments on and has incorporated into Imatest. In modern imaging using MTF, it is not sufficient to state resolution without also specifying the contrast. For example, a diffraction limited lens can resolve 200 lp/mm at the Rayleigh limit (around 10% contrast), but this contrast is too low for terrestrial imaging. Resolution at the more useful 50% contrast is only 97 lp/mm and this drops to 40 lp/mm at 80% contrast.

Just as with an aberrated camera lens, the MTF of the eye is improved by stopping down (smaller pupil size). The graph by Prof. Girod on Koren's link demonstrates that the MTF of the eye at 30 cycles per degree (60 lines/degree) is only about 17% at a pupil size of 5.8 mm but improves to around 38% at a pupil size of 2 mm. However the human visual system exhibits maximal spatial contrast sensitivity at about 6 cycles/degree.

Since the visual response is so complicated, optimal output sharpening parameters are best determined empirically, and I understand that Bruce Fraser printed thousands of images to determine the best approach, which was incorporated into PKSharpener.


Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1433


WWW
« Reply #24 on: January 07, 2010, 02:12:51 PM »
ReplyReply

Quote from: bjanes
Bart,

I'm glad to see you on LuLa. Your ImagingTechnology site is excellent, but unfortunately the forum is not that active.

Jeff's own book states (page 84), "...the secret is to keep the size of the [sharpening] halos below the threshold of visual acuity at the intended viewing distance--this is where the size of the pixels on the output becomes a critical factor"

He then goes on to say that for smaller reproductions such as used in his book (no larger than 5 x 7 inches), he tries to keep the sharpening halos to 0.01 inch, whereas for larger reproductions one can go up to 0.02 inch. If one were viewing a larger print at the same distance as for the smaller one, then it would seem as one should use the same sharpening parameters as for the smaller print. One could cut away the peripheral portion of the larger print, as it would be outside the eye's field of sharp vision in any event.

The link to Norman Koren's site regarding spatial contrast sensitivity is most appropriate for further discussion. Although the eye can resolve 60 lines per degree of arc, the frequencies around 6 cycles per degree contribute most to perceived image quality and this is the basis of the SQF measurements that Mr. Koren comments on and has incorporated into Imatest. In modern imaging using MTF, it is not sufficient to state resolution without also specifying the contrast. For example, a diffraction limited lens can resolve 200 lp/mm at the Rayleigh limit (around 10% contrast), but this contrast is too low for terrestrial imaging. Resolution at the more useful 50% contrast is only 97 lp/mm and this drops to 40 lp/mm at 80% contrast.

Just as with an aberrated camera lens, the MTF of the eye is improved by stopping down (smaller pupil size). The graph by Prof. Girod on Koren's link demonstrates that the MTF of the eye at 30 cycles per degree (60 lines/degree) is only about 17% at a pupil size of 5.8 mm but improves to around 38% at a pupil size of 2 mm. However the human visual system exhibits maximal spatial contrast sensitivity at about 6 cycles/degree.

Since the visual response is so complicated, optimal output sharpening parameters are best determined empirically, and I understand that Bruce Fraser printed thousands of images to determine the best approach, which was incorporated into PKSharpener.

From that could we assume that contrast then be a combination of light brightness, and arc perception , ultimately contrast ratio without flare etc?
Local contrast and colour contrast then too , do they play into perceived sharpeness?


Interesting thread here>   I am learning a lot.




Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2882



« Reply #25 on: January 07, 2010, 03:01:19 PM »
ReplyReply

Quote from: neil snape
From that could we assume that contrast then be a combination of light brightness, and arc perception , ultimately contrast ratio without flare etc?
Local contrast and colour contrast then too , do they play into perceived sharpeness?
I think that is right. MTF does not address veiling flare. Some say that some Leica lens produce such good results because of low veiling flare. Color adds contrast too.

I think that scientific measurements of image quality are useful, but we have to correlate the measurements with what we actually see. Michael's discussion with Norman Koren (Imatest author) on the latest video journal was quite illuminating. Michael said that he formerly did quite a bit of testing with the DXO suite, but gave up when he saw little correlation between what he was measuring and what he was seeing in the prints. An analogous between vacuum tube audio and transistor audio was also discussed.
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 8031


WWW
« Reply #26 on: January 07, 2010, 04:06:32 PM »
ReplyReply

Hi,

MTF is normally measured at different frequencies. I guess that low figure 5 lp/mm or 10 lp/mm actually relates to veiling flare, but I may be wrong on that:

There is an excellent discussion of MTF on Zeiss Camera Lens News.

I'd recommend downloading this image: http://www.zeiss.de/C12567A8003B8B6F/Graph...le/Image_02.jpg
And this article: http://www.smt.zeiss.com/C12567A8003B8B6F/...Kurven_2_en.pdf


The first image shows images with different MTF characteristics. The article discusses the MTF:s and the images.

Because the article is not easy to find I tried to collect pointers to it here: http://83.177.178.7/ekr/index.php/photoart...-and-perception

Best regards
Erik

Quote from: bjanes
I think that is right. MTF does not address veiling flare. Some say that some Leica lens produce such good results because of low veiling flare. Color adds contrast too.

I think that scientific measurements of image quality are useful, but we have to correlate the measurements with what we actually see. Michael's discussion with Norman Koren (Imatest author) on the latest video journal was quite illuminating. Michael said that he formerly did quite a bit of testing with the DXO suite, but gave up when he saw little correlation between what he was measuring and what he was seeing in the prints. An analogous between vacuum tube audio and transistor audio was also discussed.
Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8948


« Reply #27 on: January 07, 2010, 08:09:35 PM »
ReplyReply

Quote from: Schewe
What this will give you is the following viewing distance and resolution required table:

Viewing Distance (inches)____ Limit (inches)______Resolution (DPI)

8 _______________________ 0.00232 ______________ 428
12 ______________________ 0.00349 ______________ 286
15 ______________________ 0.00436 ______________ 229
18 ______________________ 0.00524 ______________ 191
20 ______________________ 0.00582 ______________ 172
24 ______________________ 0.00698 ______________ 143


A couple of things related to the above...these are not my numbers but Bruce Fraser's numbers from his Real World Image Sharpening book )both the original and the one I updated). But since I'm coauthor of the current edition I stand by them (I was also the guy who asked Bruce the original question of "how many DPI can our eyes see Bruce?" which he referred to as my sending him down yet another rabbit hole).

The above assumes 20/20 vision...there are a lot of people whose vision is better or worse. The older one gets, the less close focus vision most people have (and the more people who need to wear "reading glasses"). So you milage may vary...


Wow! That's spot on, Jeff. I noticed when viewing my still images on the 65" Plasma TV, that in order to appreciate the full detail on display, I had to sit no further from the screen than 2.5 metres, which is closer than I anticipated.

I planned on positioning the Chaise Longue near the opposite wall, about 5 metres away, and under the airconditioner so that the cold blast from the airconditioner immediately above will go straight over one's head.

Now this position of 5 metres away from the screen is better for viewing substandard material, DVDs and standard resolution video as opposed to HD. But it's simply too far away for the best quality material the TV is capable of displaying, which also includes most good quality material on Blu-ray discs.

My screen is simply not big enough for the planned arrangements.

Here are my calculations based on your table. Width of TV, 56.5 inches. Total horizontal resolution 1920 dots (or RGB pixels). Therefore, DPI = 1920/56.5 = 34 dpi.

If 143 dpi is required for a viewing distance of 24", then the viewing distance for 34 dpi should be 142/34x24 = 101" which is almost precisely 2.5 metres.
Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2882



« Reply #28 on: January 07, 2010, 09:58:23 PM »
ReplyReply

Quote from: ErikKaffehr
Hi,

MTF is normally measured at different frequencies. I guess that low figure 5 lp/mm or 10 lp/mm actually relates to veiling flare, but I may be wrong on that:

There is an excellent discussion of MTF on Zeiss Camera Lens News.

I'd recommend downloading this image: http://www.zeiss.de/C12567A8003B8B6F/Graph...le/Image_02.jpg
And this article: http://www.smt.zeiss.com/C12567A8003B8B6F/...Kurven_2_en.pdf


The first image shows images with different MTF characteristics. The article discusses the MTF:s and the images.

Because the article is not easy to find I tried to collect pointers to it here: http://83.177.178.7/ekr/index.php/photoart...-and-perception

Best regards
Erik

Erik,

I downloaded that article and the images after you recommended them in a previous post, but I don't see veiling glare (or flare) discussed. Part 1 (CLN_MTF_Kurven_EN.pdf) does discuss veiling glare on page 16 and I surmise that it is not measured by MTF: "MTF measurements say nothing about this macro contrast. They gauge only the correction of the lens, i.e. the small deviations of the light beams, while the macro contrast depends on the veiling glare of the lens, i.e. on the large deviations."

They do show a PSF demonstrating flare. I thought that the MTF could be derived from a Fourier transform of the PSF, so I am a bit puzzled here. It had been my impression that flare was normalized out of the MTF calculation. How do you interpret this? The articles are very good, but are somewhat dense to the layman.

Regards,

Bill
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 8031


WWW
« Reply #29 on: January 07, 2010, 10:52:50 PM »
ReplyReply

Hi,

My reasoning is that MTF is 100% at zero lp/mm is always 1, theoretically. Veiling flare would reduce contrast, so contrast would be less then 1. As I said i was not really sure about this and your reading indicates that I was wrong. I didn't mention the articles as a proof, but wanted to share them as they give good insight in how MTF shows up in real world. I'm going to reread the articles my self.

Best regards
Erik

Quote from: bjanes
Erik,

I downloaded that article and the images after you recommended them in a previous post, but I don't see veiling glare (or flare) discussed. Part 1 (CLN_MTF_Kurven_EN.pdf) does discuss veiling glare on page 16 and I surmise that it is not measured by MTF: "MTF measurements say nothing about this macro contrast. They gauge only the correction of the lens, i.e. the small deviations of the light beams, while the macro contrast depends on the veiling glare of the lens, i.e. on the large deviations."

They do show a PSF demonstrating flare. I thought that the MTF could be derived from a Fourier transform of the PSF, so I am a bit puzzled here. It had been my impression that flare was normalized out of the MTF calculation. How do you interpret this? The articles are very good, but are somewhat dense to the layman.

Regards,

Bill
Logged

BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3911


« Reply #30 on: January 08, 2010, 05:59:51 AM »
ReplyReply

Quote from: bjanes
I'm glad to see you on LuLa. Your ImagingTechnology site is excellent, but unfortunately the forum is not that active.

Hi Bill, I'm glad to be here. Just to make sure there is no misunderstanding, I'm not associated with Norman Koren's site (although we exchanged DSP info and other technicalities in the past). Norman has a great collection of knowledge available to all on his site, and he's a very nice person. His forum is just one method for maintaining contacts with mainly the imaging industry.

Quote
Jeff's own book states (page 84), "...the secret is to keep the size of the [sharpening] halos below the threshold of visual acuity at the intended viewing distance--this is where the size of the pixels on the output becomes a critical factor"

Yes, that's only too obvious (as long as you can't see it ...), although I'd prefer to avoid halos as much as possible in the first place. One cause for halos is repeated sharpening passes, combined with a poor resampling algorithm.

Quote
He then goes on to say that for smaller reproductions such as used in his book (no larger than 5 x 7 inches), he tries to keep the sharpening halos to 0.01 inch, whereas for larger reproductions one can go up to 0.02 inch. If one were viewing a larger print at the same distance as for the smaller one, then it would seem as one should use the same sharpening parameters as for the smaller print. One could cut away the peripheral portion of the larger print, as it would be outside the eye's field of sharp vision in any event.

I think that output sharpening should be steered by the characteristics of the output modality, not by tolerable halos as a guiding principle. The printing process (whether inkjet, photochemical, printing press, or whatever else) will add it's particular loss of detail (loss of MTF response). It's that loss that we strive to pre-compensate for. Compensating for lack of capture resolution is another subject, although in a parametrised workflow it can take place together with the final resampling and output sharpening, while avoiding cumulative artifact amplification.  

Quote
The link to Norman Koren's site regarding spatial contrast sensitivity is most appropriate for further discussion. Although the eye can resolve 60 lines per degree of arc, the frequencies around 6 cycles per degree contribute most to perceived image quality and this is the basis of the SQF measurements that Mr. Koren comments on and has incorporated into Imatest. In modern imaging using MTF, it is not sufficient to state resolution without also specifying the contrast.

Exactly, and the notion of contrast sensitivity also leads to useful remedies to compensate for output losses, or even augment the shortcomings of the capture chain. In postprocessing we can manipulate the MTF of the final image around certain spatial frequencies which can result in a higher percieved resolution (without artifacts), as long as we address the correct spatial frequencies, the ones that matter. Viewing distance does matter and, although we cannot always cater for every possible scenario, we can optimize our final MTF for the most probable situations.

Quote
Just as with an aberrated camera lens, the MTF of the eye is improved by stopping down (smaller pupil size). The graph by Prof. Girod on Koren's link demonstrates that the MTF of the eye at 30 cycles per degree (60 lines/degree) is only about 17% at a pupil size of 5.8 mm but improves to around 38% at a pupil size of 2 mm. However the human visual system exhibits maximal spatial contrast sensitivity at about 6 cycles/degree.

Yes, this tells us not only that the viewing conditions matter, but also where the most impact of our postprocessing can be expected. It also tells us that sticking to a PPI centered compensation only, for sharpening losses, is somewhat problematic to say the least, because it also doesn't recognise the impact of viewing distances. BTW personally I focus on the 8 cycles/degree eye contrast criterion, also because there are people with better than average eyesight.

Quote
Since the visual response is so complicated, optimal output sharpening parameters are best determined empirically, and I understand that Bruce Fraser printed thousands of images to determine the best approach, which was incorporated into PKSharpener.

Although I don't dismiss empirical determination, I prefer to use that as a last resort. The problem with empirically derived solutions is that we can't learn about the underlying processes, and thus potentially overlook the important principles (the ones we can use to our advantage).

Cheers,
Bart
Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3911


« Reply #31 on: January 08, 2010, 09:49:11 AM »
ReplyReply

Quote from: ErikKaffehr
MTF is normally measured at different frequencies. I guess that low figure 5 lp/mm or 10 lp/mm actually relates to veiling flare, but I may be wrong on that:

Hi Erik,

Veiling flare is primarily a localized phenomenon in an image, although in a mediocre lens design it can spread throughout the image. It will manifest itself near the lower spatial frequencies in the image capture, but it is only responsible for a (small) part of the (reduction of the) capture frequency response.

One should also be careful in not confusing the eye's MTF curve with the image's MTF. A peak contrast sensitivity at around 8 cycles/degree (~16 lines/degree) deals with print detail at a viewing distance of 8-10 inches of around 100 PPI.

Cheers,
Bart
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8948


« Reply #32 on: January 08, 2010, 11:23:44 PM »
ReplyReply

Quote from: BartvanderWolf
Although I don't dismiss empirical determination, I prefer to use that as a last resort. The problem with empirically derived solutions is that we can't learn about the underlying processes, and thus potentially overlook the important principles (the ones we can use to our advantage).

Empiricism and theory should go hand in hand. If there's a conflict, I'll opt for the empricism.

It's always been good advice, when preparing images for large prints, to increase local contrast dramatically. By that, I mean, using Photoshop Unsharp Mask settings at something like, 50 pixels radius setting and 30-50% amount, at the same time, of course, protecting any highlights and shadows which may be near to clipping.

This in effect increases the MTF of the very low frequencies, which is what you need when viewing images from a distance.
Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3911


« Reply #33 on: January 09, 2010, 04:56:03 AM »
ReplyReply

Quote from: Ray
Empiricism and theory should go hand in hand. If there's a conflict, I'll opt for the empricism.

It's always been good advice, when preparing images for large prints, to increase local contrast dramatically. By that, I mean, using Photoshop Unsharp Mask settings at something like, 50 pixels radius setting and 30-50% amount, at the same time, of course, protecting any highlights and shadows which may be near to clipping.

This in effect increases the MTF of the very low frequencies, which is what you need when viewing images from a distance.

Hi Ray,

This is an exellent example of where empiricism can lead people on a different path than the one they intended to go. A large radius unsharp mask does a local tonemapping operation, it changes the tonality based on local brightness. It does not necessarily change the MTF in a way that emphasizes targeted spatial frequencies. For that, one needs to apply a high pass filtered layer, e.g. in (although not perfect) Soft light blending mode, which is what theory tells us.

I'm not saying that a local tonemapping operation doesn't help images, it's just that it addresses another issue.

Cheers,
Bart
Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1433


WWW
« Reply #34 on: January 09, 2010, 06:07:55 AM »
ReplyReply

Quote from: BartvanderWolf
Hi Ray,

This is an exellent example of where empiricism can lead people on a different path than the one they intended to go. A large radius unsharp mask does a local tonemapping operation, it changes the tonality based on local brightness. It does not necessarily change the MTF in a way that emphasizes targeted spatial frequencies. For that, one needs to apply a high pass filtered layer, e.g. in (although not perfect) Soft light blending mode, which is what theory tells us.

I'm not saying that a local tonemapping operation doesn't help images, it's just that it addresses another issue.

Cheers,
Bart


Bart is correct.
If the 30-50% is really pixel % then what default. Or did it really refer to pixel grid width. Then 30-50 pixels  will cause a hefty shift . In that case I think you'd want to get into the habit of smart sharpening which is the way we used to sharpen when scanning. Funny how long it took to make it's way into Photoshop!

There is a lot of rage about High Pass, or high pass with a mode layer over a mid grey.
It is interesting as an effect but it sure looks like a hack to me. When printing with an enlarger without mask films, there isn't any sharpening. Still look more photographic than any high pass filter, so I am not convinced that a heavy hand is going to make a print better. Theory is one thing, but appreciation of the culture of traditional analogue imagery does not necessarily retain integrity IMO when too much of the printers hand becomes apparent.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8948


« Reply #35 on: January 09, 2010, 08:05:44 AM »
ReplyReply

Quote from: BartvanderWolf
Hi Ray,

It does not necessarily change the MTF in a way that emphasizes targeted spatial frequencies. For that, one needs to apply a high pass filtered layer, e.g. in (although not perfect) Soft light blending mode, which is what theory tells us.

Hi Bart,
Did you forget I mentioned protecting the highlights and shadows?

I always opt for the simpler system. If you are going to increase local contrast, then it has to be local, or targeted, otherwise it's global.

The simplest way of doing this is to select with the magic wand the highlights and shadows that won't benefit from contrast enhancement, feather 2 or 3 or 5 pixels, then invert.
« Last Edit: January 09, 2010, 08:06:42 AM by Ray » Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3911


« Reply #36 on: January 09, 2010, 09:35:23 AM »
ReplyReply

Quote from: Ray
Hi Bart,
Did you forget I mentioned protecting the highlights and shadows?

Hi Ray,

No I didn't, but it has nothing to do with adjusting for specific spatial frequency response. Of course one should protect shadows and highlights when applying local contrast enhancement.

Quote
I always opt for the simpler system. If you are going to increase local contrast, then it has to be local, or targeted, otherwise it's global.

The simplest way of doing this is to select with the magic wand the highlights and shadows that won't benefit from contrast enhancement, feather 2 or 3 or 5 pixels, then invert.

The Blend-If functionality in Photoshop is another efficient way of doing it, but there are many other methods possible as well.

What I am saying is that we can exploit the contrast sensitivity of the human eye, with a peak at say 8 cycles/degree, by boosting the MTF response at (and beyond) those frequencies. That has little to do with normal image contrast, it's confined to certain specific spatial frequencies and that is linked to viewing distance.

Here's a little test to illustrate:


Look at that image when you walk away from your display. At different distances you'll see a maximum contrast at different particular spatial frequencies, allowing you to discriminate lower contrast detail.

Cheers,
Bart
« Last Edit: January 09, 2010, 10:37:25 AM by BartvanderWolf » Logged
madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #37 on: January 09, 2010, 12:16:05 PM »
ReplyReply

In general you cannot optimize a single print detail-wise for all viewing distances. If you prepare an image so it looks perfect at a few inches away (sharp, natural, no artifacts) then it might look fine much farther away, but it won't be as sharp as it could be. That is, you could sharpen the image a lot more if you knew you'd be viewing it from no closer than 10 feet away. But then that same image would look bad when viewed at 10 inches.
Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8948


« Reply #38 on: January 09, 2010, 06:41:28 PM »
ReplyReply

Quote from: BartvanderWolf
Hi Ray,

No I didn't, but it has nothing to do with adjusting for specific spatial frequency response. Of course one should protect shadows and highlights when applying local contrast enhancement.

Hi Bart,
Doesn't the selection of pixel radius in the unsharp mask target the spacial frequency you want to sharpen?

For example, playing around with the following chart of really low frequencies demonstrating some very poor MTF, applying 100% sharpening with a pixel radius of 1 does nothing to increase the MTF of those frequencies. However, increasing the pixel radius to 60 produces a very dramatic improvement.

[attachment=19305:Comparison_of_3.jpg]

Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3911


« Reply #39 on: January 10, 2010, 10:04:20 AM »
ReplyReply

Quote from: Ray
Doesn't the selection of pixel radius in the unsharp mask target the spacial frequency you want to sharpen?

Hi Ray,

Yes, pixel radius does, but that's only true for the radius that is used and frequencies in that neighborhood (especially somewhat lower frequencies).

Quote
For example, playing around with the following chart of really low frequencies demonstrating some very poor MTF, applying 100% sharpening with a pixel radius of 1 does nothing to increase the MTF of those frequencies. However, increasing the pixel radius to 60 produces a very dramatic improvement.

Be careful with mixing up spatial frequencies in the image, with spatial frequencies in the MTF of our eyes. Also, the large radius sharpening works out differently on areas of different large area brightness distribution, but hardly addresses spatial frequencies (other than overall brightening/darkening) near the limiting resolution of the image.

Cheers,
Bart
Logged
Pages: « 1 [2] 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad