Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: A useful sharpening trick?  (Read 1737 times)
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7647


WWW
« on: April 02, 2013, 01:14:18 PM »
ReplyReply

Hi,

I would suggest a small sharpening trick. I would call it two step sharpening.


I normally capture sharpen in Lightroom, using:

Amount45
Radius0.7
Detail100
Masking17
Noise reduction20

These setting enhance fine detail optimally, according to my taste. What I discovered that the image can be improved a lot by doing an additional step of sharpening with a large radius and a small amount. I would use unsharp mask in photoshop and sharpen with radius = 3 and 25%.

The effect is quite subtle but quite obvious if applied in PS over a layer that is toggled.

As a side note, the detail slider operates so that at left position it supresses halos while in the right position it switches to deconvolution. I normally use radius 0.7 for sharp images, 1.0 for images that are less optimal and 1.3 to compensate for diffraction at f/16.

Best regards
Erik
« Last Edit: April 02, 2013, 01:45:20 PM by ErikKaffehr » Logged

jerryrock
Sr. Member
****
Offline Offline

Posts: 565



WWW
« Reply #1 on: April 02, 2013, 02:20:42 PM »
ReplyReply

I found that capture sharpening is sensor specific. The sharpening formula I use for the Canon EOS 7D is not the ideal formula for the Canon EOS 1Ds.
Logged

Gerald J Skrocki
skrockidesign.com
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7647


WWW
« Reply #2 on: April 02, 2013, 02:34:45 PM »
ReplyReply

Hi,

I absolutely agree. On the other hand the settings I use may be worth a try as a starting point.

Also, what I found and what I think may be more universal is that an additional step of low amplitude sharpening enhances the image. It is of course possible sharpen with a moderate amount at a large radius, but that will not improve fine detail.

Best regards
Erik

I found that capture sharpening is sensor specific. The sharpening formula I use for the Canon EOS 7D is not the ideal formula for the Canon EOS 1Ds.
Logged

jrsforums
Sr. Member
****
Offline Offline

Posts: 744


« Reply #3 on: April 02, 2013, 03:06:06 PM »
ReplyReply

Hi,

I absolutely agree. On the other hand the settings I use may be worth a try as a starting point.

Also, what I found and what I think may be more universal is that an additional step of low amplitude sharpening enhances the image. It is of course possible sharpen with a moderate amount at a large radius, but that will not improve fine detail.

Best regards
Erik


Erik, could you provide more detail on the two points you made above?

Thanks, John
Logged

John
fike
Sr. Member
****
Offline Offline

Posts: 1373


Hiker Photographer


WWW
« Reply #4 on: April 02, 2013, 03:15:58 PM »
ReplyReply

I also agree that the capture sharpening depends on the camera.  I do completely different sharpening routines on my OM-D than I do on my 7D.

As for that large radius low percentage sharpening...I find that type of sharpening approach helps to cut haze and is a sort local contrast enhancement. It functions similarly to the use of a new layer with a highpass filter and an overlay blending mode. 
Logged

Fike, Trailpixie, or Marc Shaffer
marcshaffer.net
TrailPixie.net

I carry an M43 ILC, a couple of good lenses, and a tripod.
Isaac
Sr. Member
****
Offline Offline

Posts: 2890


« Reply #5 on: April 02, 2013, 04:40:11 PM »
ReplyReply

and is a sort local contrast enhancement.

So given they're talking about LR, like clarity?
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7647


WWW
« Reply #6 on: April 02, 2013, 04:43:49 PM »
ReplyReply

Hi,

I made the TIFFs and measured MTF with Imatest.

Left image: My default sharpening
Center: My defaults + 15% unsharp mask at r=3
Right: Coarse sharpening in LR where I used my defaults but increased radius to 3 in LR

Now, check MTF at 60 lp/mm:
Left image: 0.8
Center image: 0.95
Right image: 0.55

This was a very coarse demo, but I need some sleep ;-)

The problem with the left method is that it doesn't push the MTF for low frequencies where the eye is most sensitive. The righthand method does not really improve that much on fine detail, but it was just a coarse demo.

Note: This test image was made with a 100/2.8 Macro at optimum aperture, sharper than usual.

Best regards
Erik



Erik, could you provide more detail on the two points you made above?

Thanks, John
« Last Edit: April 02, 2013, 05:08:25 PM by ErikKaffehr » Logged

Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad