Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
08-23-2021, 12:24 AM - 2 Likes   #1
Pentaxian




Join Date: Feb 2015
Photos: Gallery
Posts: 12,173
Resolution # Sharpness

Interesting finding. Resolution and sharpness (acutance) are two different things. Higher resolution display/print prevents images to look sharp. Sharpening has little visual effect on high resolution images, and is much more noticeable on lower resolution images. That observation leads me to believe I was accustomed to the look of over-sharpened images, whereas high resolution images show smooth tonal gradations with or without sharpening. The baseline for sharpening image, when enlarging them, should be based on the look of a high resolution image (insensitive to pixel sharpening). I now believe most digital images have been sharpened too much.

08-23-2021, 01:17 AM - 1 Like   #2
Site Supporter
Site Supporter




Join Date: Jan 2015
Location: Sweden
Posts: 338
Certainly! A photo can never have too high resolution (positive) - but very often photos are over-sharpened (negative) during post processing. There also seem to be a "general" confusion about these, and many clump resolution and sharpness into one.

I personally always prioritize resolution (f .ex. choosing and testing lenses), it build strong photos that handles well and look good.
08-23-2021, 03:52 AM - 1 Like   #3
GUB
Loyal Site Supporter
Loyal Site Supporter
GUB's Avatar

Join Date: Aug 2012
Location: Wanganui
Photos: Gallery | Albums
Posts: 5,735
QuoteOriginally posted by Stefan Jr Quote
Certainly! A photo can never have too high resolution (positive) - but very often photos are over-sharpened (negative) during post processing. There also seem to be a "general" confusion about these, and many clump resolution and sharpness into one.

I personally always prioritize resolution (f .ex. choosing and testing lenses), it build strong photos that handles well and look good.
A bit of a generalisation here.
If an image, when viewed looks sharp, why should it matter what resolution it is ? The obvious exception is of course when the pixels are visible.
An image after all is about the viewing of it.
08-23-2021, 04:16 AM - 1 Like   #4
Pentaxian




Join Date: Feb 2015
Photos: Gallery
Posts: 12,173
Original Poster
QuoteOriginally posted by GUB Quote
If an image, when viewed looks sharp, why should it matter what resolution it is ?
It should matter because the impression of sharpness may come from aliasing: insufficient resolution, and too much sharpening to compensate for the lower resolution, doesn't render the same natural look as the image with higher pixel density.

08-23-2021, 04:34 AM - 2 Likes   #5
GUB
Loyal Site Supporter
Loyal Site Supporter
GUB's Avatar

Join Date: Aug 2012
Location: Wanganui
Photos: Gallery | Albums
Posts: 5,735
QuoteOriginally posted by biz-engineer Quote
It should matter because the impression of sharpness may come from aliasing: insufficient resolution, and too much sharpening to compensate for the lower resolution, doesn't render the same natural look as the image with higher pixel density.
A personal observation but I don't believe the human eye/ brain combo observes resolution much at all. If we did then the impressionists would have failed miserably. But "sharpness" perception is highly tuned - I feel we take as many clues from it as we do from our binocular vision to gauge depth in an image.
EDIT or more to the point because we can't utilise our binocular vision in an image we are then highly reliant on relative sharpness to define depth. (and of course distance colour).
08-23-2021, 04:47 AM - 2 Likes   #6
Moderator
Loyal Site Supporter
Loyal Site Supporter
pschlute's Avatar

Join Date: Mar 2007
Location: Surrey, UK
Photos: Gallery
Posts: 8,110
My understanding has always been that the sharpest image will be obtained when it has both high resolution and high acutance (edge contrast transition).

A sharpened low-resolution image leads to the "blocky" look.

An unsharpened high resolution image looks much smoother but can have a "soft" apearance.

A high resolution image with high acutance will produce the sharpest image.

https://www.cambridgeincolour.com/tutorials/sharpness.htm
08-23-2021, 04:47 AM   #7
Pentaxian




Join Date: Feb 2015
Photos: Gallery
Posts: 12,173
Original Poster
QuoteOriginally posted by GUB Quote
I don't believe the human eye/ brain combo observes resolution much at all.
The brain gets fooled and I discovering what the right amount of sharpening should look like, i.e natural look instead of crispy look. Typically, increases resolution tends to smooth out excess sharpening. 300ppi doesn't look crisp, but it can show more details looking close. 125ppi may look crisp, not further detail revealed when looking close.

08-23-2021, 06:12 AM - 4 Likes   #8
Pentaxian
photoptimist's Avatar

Join Date: Jul 2016
Photos: Albums
Posts: 5,113
Technical definitions of resolution center on the yes-or-no ability to distinguish or resolve small features in the image. Standard resolution test targets call for determining if one sees one thick line or two separate thin lines in a close-pair of lines. In astronomy, telescope resolution depends on determining if one sees one star or two stars in a close pair.

Thus, resolution assesses dark-light-dark-light-dark patterns of the repeated lines of a test target or a potential pair of stars in a dark sky.

"Sharpness" or acutance really doesn't have a good technical definition but is more a perception of the width and strength of single dark-light transitions at the edges of objects in the scene such as the edge of a leaf, building, or letter of text. Are the edges between black and white subjects fuzzy (black-darkgray-gray-lightgray-white) or sharp (black-black-white-white-white)?

Opportunities for enhancing resolution in post are limited by the original data -- if there's not a distinct dark pixel/patch between the two light pixel/patches in the original data (that exceeds the noise-levels of the image), there's nothing to enhance.

Opportunities for enhancing sharpness or acutance in post are more open-ended -- any multi-pixel transition between light and dark can be narrowed or intensified to make the transition look sharper.
08-23-2021, 07:08 AM   #9
Pentaxian




Join Date: Feb 2015
Photos: Gallery
Posts: 12,173
Original Poster
QuoteOriginally posted by photoptimist Quote
Opportunities for enhancing sharpness or acutance in post are more open-ended -- any multi-pixel transition between light and dark can be narrowed or intensified to make the transition look sharper.
Single pixel or multi-pixel sharpening. Sharpening (accentuation) is done according to a certain radius. If I export two versions of the same image: one sharpened with a radius of one pixel, one not sharpened, I display at 140ppi, my eyes don't see much difference between the sharpened and non sharpened images. That happens because the resolution of the display exceed what my eyes can see, my eyes don't see the effect of pixel sharpening and don't even see that the noise increased due to pixel sharpening. I previously sharpened with an HD display (~62ppi) and felt that the prints lacked sharpness! Now I know why. The sharpening radius should be set according print ppi. If the display doesn't have the same resolution as the print, it's hard to judge how much sharpening and radius should be used. Can I apply a ratio to radius/amount , based on the ratio of display resolution/print resolution, so that to achieve the same look on display and in print?

Last edited by biz-engineer; 08-23-2021 at 07:14 AM.
08-23-2021, 09:42 AM - 1 Like   #10
Pentaxian




Join Date: May 2016
Photos: Gallery
Posts: 3,722
QuoteOriginally posted by biz-engineer Quote
Can I apply a ratio to radius/amount , based on the ratio of display resolution/print resolution, so that to achieve the same look on display and in print?
I think that's the key: sharpening algorithms work on pixel counts (radius etc), but we view images in terms of spatial resolution (dpi, lines per inch, whatever). That's why one can/should sharpen on the resolution for printing, viewed at the dpi the screen/print will look like. Otherwise if I take a 36mp (or whatever) image, sharpen it, then downsize it to 2mp so it has 300dpi for a 4x6 print, whatever effects were on the 36mp are lost due to down sampling. As pixel density, the 36mp image would be technically able to produce 1200 dpi for 4x6 print size, though I think we'd need a microscope at that point to see any more detail.

btw this was a great thread to read so far on sharpness vs actuance, something to think about when processing even if I'm not printing

Last edited by aaacb; 08-23-2021 at 09:48 AM.
08-23-2021, 09:53 AM - 2 Likes   #11
Pentaxian
photoptimist's Avatar

Join Date: Jul 2016
Photos: Albums
Posts: 5,113
QuoteOriginally posted by biz-engineer Quote
Single pixel or multi-pixel sharpening. Sharpening (accentuation) is done according to a certain radius. If I export two versions of the same image: one sharpened with a radius of one pixel, one not sharpened, I display at 140ppi, my eyes don't see much difference between the sharpened and non sharpened images. That happens because the resolution of the display exceed what my eyes can see, my eyes don't see the effect of pixel sharpening and don't even see that the noise increased due to pixel sharpening. I previously sharpened with an HD display (~62ppi) and felt that the prints lacked sharpness! Now I know why. The sharpening radius should be set according print ppi. If the display doesn't have the same resolution as the print, it's hard to judge how much sharpening and radius should be used. Can I apply a ratio to radius/amount , based on the ratio of display resolution/print resolution, so that to achieve the same look on display and in print?
Yes, the best sharpening parameters depend on the display or printing parameters relative to the viewing conditions and the human eye.

The other issue is that the sharpening parameters depend on the blur properties of the optics and camera. If, for example, the blur created by the lens, camera motion, low-pass-filter, and Bayer pixel pattern totals 3 pixels in radius, a 1 pixel radius sharpening won't actually sharpen the image very much. You can think about the issue in the frequency domain, too. If the lens-camera-system has attenuated response of features that are 6 pixels/cycle, the correction needs to amplify the 6 pixels/cycle part of the signal.
08-23-2021, 11:37 AM - 2 Likes   #12
Loyal Site Supporter
Loyal Site Supporter




Join Date: May 2007
Location: Flagstaff, Arizona
Posts: 1,625
Somewhat relevant, but perhaps more than you want to know (from the folks at Imatest): Comparing sharpness in cameras with different pixel count | imatest
08-23-2021, 01:20 PM   #13
Pentaxian




Join Date: May 2015
Photos: Albums
Posts: 3,304
One curious thing is that small sensor output often look over sharpened whilst large sensor output look smooth and detailed. The curious thing is that this seems to be the case even when you down sample both files. It's hard to detect reliably and consistently but it's one of the advantages I've noticed with going FF even for deep dof web size images.

Could be the processing but since I see it in my own images I doubt it.
08-23-2021, 11:35 PM   #14
Pentaxian




Join Date: Feb 2015
Photos: Gallery
Posts: 12,173
Original Poster
QuoteOriginally posted by AstroDave Quote
Somewhat relevant, but perhaps more than you want to know (from the folks at Imatest): Comparing sharpness in cameras with different pixel count | imatest
Great link thanks.

---------- Post added 24-08-21 at 08:41 ----------

QuoteOriginally posted by photoptimist Quote
You can think about the issue in the frequency domain, too. If the lens-camera-system has attenuated response of features that are 6 pixels/cycle, the correction needs to amplify the 6 pixels/cycle part of the signal.
Yep, thanks. To compensate for the lens, I'll eventually take a picture of something with sharp details (or a camera test target) so that I can tweak my sharpening setting for realistic look.
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
image, images, imatest, look, photography, photoshop, resolution, sharpness, thanks
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Sharpness vs Fine Sharpness don_park Pentax K-3 & K-3 II 5 03-20-2016 07:45 PM
Sharpness and fine sharpness settings Altglas Pentax Q 7 08-03-2013 01:39 PM
Sharpness, Fine Sharpness, Extra Sharpness Taviali Pentax K-5 & K-5 II 8 03-31-2012 12:56 PM
k-7 sharpness / fine sharpness / fine sharpness 2 mattdm Pentax DSLR Discussion 10 12-12-2010 08:10 AM
Sharpness? Resolution? Detail? Contrast? edumad Photographic Technique 6 04-14-2010 01:33 AM



All times are GMT -7. The time now is 06:21 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top