On 7/7/2012 5:41 PM, Tina Manley wrote:
> Do you have a tutorial somewhere on how to do that? I would love to know.
Are you a baker? I'm a pretty good cook, but not much of a baker. Years ago, I
decided to learn to make a good pie
crust. After a couple of far less than successful efforts, I did some research.
It appears that the recipe is just a
rough guideline; the good baker adjusts based on the particular flour,
temperature, humidity, perhaps even phase of the
moon, for all I know. I still buy pie crusts. :-)
Using PS, beyond the simple things one may as easily do on LR, Aperture, ACR,
etc., is is some ways like baking. There
may be a recipe, but one needs to be able to adapt to particular circumstances.
This is particularly true here, as you would be working on the original size,
while I was working with a much reduced
size. The settings would be different.
I use NeatImage for noise and grain reduction. I suppose other apps may work as
well or better, but it's both what I
have and what I know.
I've made a NI noise profile for your TMax 3200, as scanned on your Nikon, and
I used that.
Then I made two layers of different amounts of NR. The first reduces only high
frequency noise, with a little
resharpening. I visually tuned it for the face and arm of the central woman. It
nicely reduces the high freq. noise
without losing detail.
The darker parts, particularly lower left, have almost no image detail, but a
lot of uneven blotchiness. Whether from
the film itself or scanning, I don't know, but it's common in deep shadows of
your TMax 3200 images. Applying lots of
mid and low frequency NR removes most of that problem.
One may, of course, fine tune the NR effects by changing the Opacity of each
layer.
So the final image mixes three levels of NR. In addition to the two levels of
NR described above, the tortillas are
unmodified. They are bright enough that there is little visible noise - and the
noise looks a lot like uneven browning
and natural texture. NR just makes them go blah.
The question then is how to mix the other two NRs. I could simply paint on a
mask layer, as effects are applied in a lot
of editors, but I like something more precise, particularly in an image like
this, with a complex mix of light and dark.
One good way to make such a mask is using Select=> Color Range and either
Select Highlights or Select Shadows (I have
both set as Actions.) I first tried Select Shadows, but it selected too much. I
used Brightness (Again, with an Action
that creates a new layer and opens the dialog box.), about 60, I think, in a
new temporary layer, to be used only for
selection, then Select Shadows. This selection looked pretty good to me. Then
I selected the strong NR layer and
applied the selection as a layer mask.
As you can see in the third image here, the mask is very sharp edged. Sometimes
that can look funny. I also did just the
same thing except that I used Refine Edge on the selection to soften the edges,
as in the fourth image. The way these
masks work is that white areas are visible, black areas are invisible, showing
through what's below the layer, and
shades of gray are partially opaque.
In this case, I didn't see any difference between the two masks, but show both
as you asked about technique.
All this takes less time than you might think, as I have Actions to speed up
many parts of the process and have done
such things many, many times. Still it's not something to contemplate doing to
every B&W scan. :-)
There's a lot of semi-artistic judgement involved. I could, for example, have
completely eliminated the larger scale
unevenness in the lower left with NR and/or darkening it. But I thought it
looked rather like weak light on an uneven
dirt floor, which it may be.
Grainless Moose
--
What if the Hokey Pokey *IS* what it's all about?
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|