Digital noise 73power tool, such as Focus Magic (www.focusmagic.com), that
can handle both problems.
When working with a noisy image that is also blurred, Focus
Magic looks through the noise and sees the underlying image.
It does this by knowing that when an image is out of focus,
the image can have only gentle gradients—and the more out of
focus the image, is the gentler the gradients must be. Using
this information, Focus Magic differentiates between noise
and image. The amount of noise removed is displayed at
the bottom of the dialog box each time you refocus a small test
area on the image, and this noise is removed before the image
is sharpened.
The noisiness of the image is taken into account during the
sharpening of the image. The Image Source control determines
how aggressively or conservatively the focusing algorithm
should be in deconvoluting the blur, and takes into account the
nonlinear properties of the capturing device, whether it’s a CCD/
CMOS sensor for a digital camera or fi lm. For nongeeks, decon-
volution is an image-processing fi lter that uses Fourier transform
mathematics to restore a blurred image as nearly as possible to
an unblurred state.
Although Focus Magic isn’t designed to be a noise-removal tool,
you can use it to remove noise, only as follows:If you run a noisy image through
a normal Unsharp Mask–based
sharpening tool, noise increases
signifi cantly. That’s when you
need a power tool, such as Focus
Magic, that can handle both prob-
lems. © 2004 Joe Farace.