Advances in Continuous and Discrete Models

Theory and Modern Applications

Table 2 Quantitative analysis of obtained results of FPMM and PMM for denoising the test case 2 with the Gaussian noise

From: A time-splitting local meshfree approach for time-fractional anisotropic diffusion equation: application in image denoising

	SNR	PSNR	SSIM	MSE
Noisy image (variance 0.01)	15.0030	20.4392	0.3417	587.7050
FPMM (α = 0.9, K = 90, iteration = 20)	20.1193	25.7335	0.6025	203.2346
PMM (K = 100, iteration = 20)	17.8155	23.3781	0.5845	190.7179
Noisy image (variance 0.03)	10.6852	15.8952	0.2126	1.67e + 03
FPMM (α = 0.9, K = 110, iteration = 42)	16.8226	22.4020	0.4141	419.4402
PMM (K = 100, iteration = 42)	16.3136	21.8572	0.3919	431.1948
Noisy image (variance 0.05)	8.8399	13.8739	0.1647	2.66e + 03
FPMM (α = 0.9, K = 110, iteration = 70)	14.8705	20.4323	0.3384	654.0687
PMM (K = 110, iteration = 70)	14.4288	19.9599	0.3143	668.8500

Back to article page