You've answered a question I hadn't yet imagined :-)
Well, this is what's been puzzling me:
I've been doing USM and LCE the way I learned here.
After you posted the link
<http://www.luminous-landscape.com/tutorials/high-pass-sharpening.shtml>,
I went a little further and found something which seemed obvious:
making the sharpening in the luminance channel only - maybe everyone
did so and I hadn't noticed.
Take into account I use film only, and my 'noise' standard is that of
the 4000ED - nonetheless either DFine or NeatImage 'find' it noisy,
specially if I play with NikonScan curves to balance color and contrast,
or use Nikon Scan UnsharpMask.
Found this very clever
<http://www.luminous-landscape.com/tutorials/instant_photoshop.shtml>:
"Go to *Image / Mode / Lab Color* and then *Window / Show Channels*. In
the *Channels *palette select the *Lightness channel*. Now choose
*Filter / Sharpen / Unsharp Mask*."
_ and I felt happy for a couple of days, until found this thread:
<http://www.luminous-landscape.com/forum/lofiversion/index.php/t28753.html>
Wow - if there's trouble when going to sRGB from Adobe RGB '98 (
"conversion *intents*" ) - it should be worse when going from RGB to
Lab, and coming back to RGB !!!.
Thanks to Ken, I had scanned another Velvia 50 cactus and nailed in the
green, but sharpened in Lab. Found nothing too wrong, ignorance still
bliss, I hadn't found that thread yet. Maybe you could make the effort
and glimpse for a second or two this huge Flickr file - I won't feel bad
if you refuse:
small: <http://www.flickr.com/photos/fernando_gonzalez_gentile/3096675514/>
large:
<http://www.flickr.com/photos/fernando_gonzalez_gentile/3096675514/sizes/o/>
(tripod mounted 65-116, 135/4,5 @8, maybe 1:10x, OM 2n on Auto mode
+1/3, Velvia 50, perceptual intent conversion to sRGB, NeatImage).
OK, despite it looks right for me, the rationale of a round trip to Lab
is confusing. I don't completely understand Lab, and feel too lazy to
read Dan Margulis books (http://en.wikipedia.org/wiki/Dan_Margulis).
Months ago, Chuck posted another tutorial - which I found one of the
most useful:
Found in <http://www.cambridgeincolour.com/tutorials/unsharp-mask.htm>
this advice:
_Remedies_: Color shifts can be avoided entirely by performing the
unsharp mask within the "lightness" channel in LAB mode.
**A better technique, which avoids converting between _color spaces
<http://www.cambridgeincolour.com/tutorials/color-spaces.htm>_ and
minimizes posterization
<http://www.cambridgeincolour.com/tutorials/posterization.htm>, is to**:
1) Create a duplicate layer
2) Sharpen this layer like normal using the unsharp mask
3) Blend sharpened layer with the original using "luminosity" mode in
the layers window
Tried it two days ago on a .jpg out my 720SW - it works nice too.
What I now feel reluctant to use is the traditional USM - I'm beared
towards 'sharpeining' in the Luminance channel - so:
1- the 'high pass' technique - is or is not related to Luminance
channel: seems so.
2- Lab .... ?
3- Blending sharpened layer using Luminosity Mode ... subtle sharpening
and have two sliders to play with, which I didn't understand yet.
Last:
1- what if I do nothing in NikonScan, except ICE, and work anything in CS3.
2- *when* in the workflow do I apply Luminance sharpening. If I have
amusing color shifts while USMasking this was a non issue ... but this
is getting more complex, isn't it ?
TIA
Fernando
Chuck Norcutt wrote:
> I hope that answered the question you haven't yet asked. :-)
>
> Chuck Norcutt
>
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|