Blind Source Parameters for Performance Evaluation of Despeckling Filters

Table 2

Types of despeckling techniques.

Anisotropic diffusion: AD [1, 31],; CED [1, 33], DPAD [4, 34], SRAD [1, 32]	SAR: Lee [1, 7, 8, 23], Kaun [1, 7, 8, 36], Frost [4, 37]	Wavelet shrinkage: BayesShrink [21, 25], OWT [21, 26], MPT [21, 27], ProbShrink [21, 30], soft thresholding [3, 6], NeighShrink [28], SURELET [21, 29]
Total variation: TV [12, 18], AFTV [16], ATV [17]	Nonlocal means: OBNLM [14], PPB [15]
Despeckling filters: DsFlsminsc, DsFlsmv, DsFhomog, DsFwiener, DsFmedian, DsFhmedian [5, 9, 12]	Fuzzy: TMED, TMAV, ATMED, TMAV, HTMAV, HTMED, HATMED, GWF [21, 24]	Multiscale techniques: GLM [1, 4, 13, 21], MBR [9, 38], RNLA [11, 12], PSBE [10]
Fast bilateral filter [12, 20]	Geometric filter [1, 5, 35]	Fourier: FIF, FBF, HFIF, HFBF [6]

MPT: multiscale product thresholding; BayesShrink: Bayes thresholding; OWT: orthogonal wavelet thresholding; SURE: Stein’s unbiased risk estimation; LET: linear expansion of threshold; NeighShrink: neighborhood shrinkage; PSBE: posterior sampling based Bayesian estimation; GLM: generalized likelihood ratio filtering method; TV: total variation.