Compensation of local contrast reduction due to bad pixel interpolation may lead to a loss in information.
Please refer to chapter for detailed information.

Hi Klaus:

Recently, I have been focusing on Compensation Principle.

In Compensation Principle III , you mentioned it's for "local unsharpness",
however in the paper, above fig.6, "The interpolation of bad pixels can reduce the contrast of very small indications ",

So it seems that compensation principle III is for contrast loss (caused by calibration of bad pixel), instead of unsharpness.

Almost all of the DR standards said compensation principle III is for "local unsharpness", so why is unsharpness?

Hi Colin,
you are right, the third compensation principle is for local contrast loss ==> see .
I corrected it in my first posting here.
Thank you for this hint

Hi Klaus:

Since CNR=C*SNR, and when the voltage increases, the C will decrease and SNR will rise.

Maybe the CNR could be rewritten as a function of the voltage, and the best CNR can be calculated by the derivative of voltage in that function.

So, has this work been done ?
I think how to set a good voltage is quite useful.