The most difficult restoration task is the one with maximum degradation and highest noise level (noise with sigma=16 pixel, maximum blurring). This example gives an qualitative and quantitative impression of how the various methods perform in a very difficult restoration task. For compactness the images and rms errors are summarized in a table.

Defocus (sigma=8 pixel)

Inverse Filter (cut-off)			Wiener Filter
0.01 (rms=67.64)	0.1 (rms=40.89)	1.0 (rms=46.14)	1.0 (rms=38.07)
Inverse Filter (freq. cut-off)			Power Spectrum Filter
10 (rms=41.38)	40 (rms=40.05)	80 (rms=40.49)	1.0 (rms=94.14)
Constrained Least Square
0.001 (rms=96.97)	0.01 (rms=50.76)	0.1 (rms=37.72)	1.0 (rms=36.77)

It can be observed very clearly that all restoration methods struggle with this task. The high noise level makes a good restoration very difficult. An unbiased person that had no knowledge of the actual original image identfied the two trees in the image but did clearly not point out a sailboat. The Wiener filter, the inverse filter and the constrained least square filter all produced comparable results.

Motion (length=64 pixel)

Inverse Filter (cut-off)			Wiener Filter
0.01 (450.54)	0.1 (46.72)	1.0 (67.17)	1.0 (40.98)
Inverse Filter (freq. cut-off)			Power Spectrum Filter
10 (46.35)	40 (49.48)	80 (57.26)	1.0 (112.20)
Constrained Least Square
0.001 (75.87)	0.01 (46.89)	0.1 (38.36)	1.0 (38.65)

The results for the motion degradation where clearly slightly better, although the degraded image did not appear to be less degraded than the images in the above case of gaussian blurr. Qualitatively the Wiener filter outperformed the other methods. With the constrained least square method a close second. This could also be observed in the rms errors.