Re: Scientifically Based Presharpening for Enlargement

jim <"sjedgingN0sp"@m@xxxxxxx> writes:

The general premise is correct. The box-car like convolution of the
sensor array has a known frequency response and a sharpening filter when
applied goes along way to reverse (it has the inverse frequency
response) the effects of the sampling. So in theory, if you start with a
appropriately bandlimited image and you carry out this exercise you
should be able to come very close to recovering the original image.

And the sharpening filter's frequency response should be matched to that
of the low-pass filter caused by the finite area of the sensor array
sensels, so that the product of the two is close to flat. That way, the
image should be closer to the original after the sharpening step.

But if you look at the two small "before interpolation" images, you can
see that the presharpening step has added halos (ringing artifacts)
around abrupt edges like telephone poles. So the sharpening step is
boosting some frequencies more than they were attenuated, not merely
compensating for the attenuation. These artifacts are then enlarged by
the resampling process.

What that means is, in order for your experiment to work, you need to
start with a high resolution blurry image (appropriately bandlimited so
that no aliasing occurs when you resample it) and then (after simulation
of capturing with a sensor array) your pre-sharpening technique will do
a better job of recovering the original blurry image than interpolation
without it.

After adjusting the sharpening filter so it merely restores losses, not
oversharpens edges until they overshoot.

Dave
.

Quantcast