Regularized Multiframe Super-Resolution Image Reconstruction Using Linear and Nonlinear Filters

Table 4

A numerical comparison of SSIM for various SR approaches.

Image	LOR	L1-BTV	BEP	L2-BTV	MMW-BTV	MLRW-BTV
Acen	0.81020	0.82641	0.85083	0.96882	0.96888	0.96913
Cartap	0.88417	0.85784	0.90570	0.97988	0.98220	0.98219
Foreman	0.88533	0.80184	0.85553	0.93580	0.93735	0.93650
Text	0.74471	0.77899	0.80922	0.96317	0.96593	0.96530
Brain	0.89743	0.86688	0.89533	0.93307	0.93347	0.93342
License Plate	0.81435	0.84267	0.86751	0.95510	0.95650	0.95640
Natural Scene	0.88567	0.87646	0.87795	0.94985	0.94987	0.94991
Parrot	0.91695	0.92255	0.92669	0.97674	0.97746	0.97731
Average SSIM	0.854851	0.846705	0.873595	0.957804	0.958958	0.95877

The bold numbers indicate the highest values of the proposed methods.