The best case is considered to be the the one where the noise level is minimal. We choose to use maximum blurring in this example in order to show the performance of the various restoration algorithms under optimal conditions. The noise that was added to the degraded images has a variance corresponding to sigma=1.0 pixel.
 

Defocus (sigma=8 pixel)

 

Inverse Filter (cut-off)			Wiener Filter
0.001 (61.43)	0.01 (32.89)	0.1 (37.56)	1.0 (32.40)
Inverse Filter (freq. cut-off)			Power Spectrum Filter
10 (38.09)	40 (35.69)	80 (35.70)	1.0 (91.01)
Constrained Least Square
0.001 (33.28)	0.01 (32.49)	0.1 (34.23)	1.0 (36.21)

Table 0: Restoration results for the various methods and varying restoration
parameters. The restoration parameters and rms errors are given below each image.

Compared to the "worst case", the results where better, but still far from the original image. All methods gave approximatelt comparable results with the exception of the power spectrum filter.

Motion (len=64 pixel):
 
 

 

Inverse Filter (cut-off)			Wiener Filter
0.001 (109.77)	0.01 (85.08)	0.1 (37.82)	1.0 (32.28)
Inverse Filter (freq. cut-off)			Power Spectrum Filter
10 (43.33)	40 (34.58)	80 (33.94)	1.0 (110.61)
Constrained Least Square
0.001 (31.39)	0.01 (32.44)	0.1 (34.92)	1.0 (38.02)

Table 0: Restoration results for the various methods and varying restoration
parameters. The restoration parameters and rms errors are given below each image.

Compared to the restoration of the gaussian blurring the restoration of the motion degraded images is far better. Especially the Wiener filter and the constrained least square method gave good results (subjectively and in terms of rms error) and an unbiased person identified the trees, the forest in the background and "guessed" a lake in the foreground.