A New Boosting Algorithm for Shrinkage Curve Learning

Figure 3

Relationship among sequence , PSNR, and iteration number. The image corresponding to this example is a fingerprint with a size of . The simulated noise is Gaussian noise with zero mean and 10 standard deviations. (a) The sequence converges rapidly to 0 when increases. (b) PSNR tends to be stable with increasing of .

(a)

(b)