As many image restoration techniques are continuing to be developed, it is increasingly difficult to compare the performance of various methods. Although some image-quality measures have been presented in the literature [1], it is inappropriate to choose a particular measure as a benchmark of performance evaluation for a wide range of applications. More importantly, none of these quality measures can be used as a performance bound which usually indicates how much potential performance can be improved for a specific restoration scheme. Therefore, it is extremely important to develop theoretical performance bounds under a variety of image and noise models for general image restoration schemes.