Edge-aware nonlinear diffusion-driven regularization model for despeckling synthetic aperture radar images

EURASIP Journal on Image and Video Processing

Table 1 No-reference image quality values based on the von Mises distribution. (Images in the first, second, and third columns refer to specific channels in the Indian pines dataset)

Methods	von Mises distribution quality value
Methods	\(\mathrm {120^{th}}\)-Channel	\(\mathrm {70^{th}}\)-Channel	\(\mathrm {50^{th}}\)-Channel	Washington	Soil Salinity	Las Vegas
Perona–Malik [26]	0.1168	0.1429	0.1292	0.3248	0.2951	0.1593
Guo [45]	0.1134	0.1133	0.1133	0.5539	0.4971	0.1445
Total variation [28]	0.1180	0.1414	0.1266	0.5017	0.3574	0.1659
Charbonnier [40]	0.1131	0.1190	0.1132	0.1799	0.1819	0.1130
DNN [46]	0.1229	0.1429	0.1292	0.5949	0.5366	0.1745
FFDNet [47]	0.3490	0.2421	0.2706	0.5751	0.4071	0.1867
Our model	0.1229	0.1443	0.1305	0.5708	0.4051	0.1745