Multiframe Superresolution Reconstruction Based on Self-Learning Method

Table 1

The results of PSNR, MAE, and SSIM. For every image, there are three rows. The first, second, and third rows for each test image indicate PSNR, MAE, and SSIM values.

	Interpolation [4]	NLM method [8]	Glassner’s method [25]	Yang’s method [23]	Zhang’s method [27]	The proposed method
Lena	27.3993	27.4644	30.6150	31.0123	31.5623	32.6016
	6.0320	5.8437	4.2831	3.9561	3.8601	3.7735
	0.8631	0.8643	0.9051	0.9168	0.9250	0.9538
Building	31.2571	31.2962	37.1000	37.0549	36.9531	37.6990
	4.1998	4.0993	2.3300	2.3254	2.5365	2.0510
	0.8746	0.8753	0.9537	0.9524	0.9124	0.9653
Flower	26.3318	25.7424	30.5731	31.2439	31.3826	32.1054
	8.0847	8.3127	5.240	4.6710	4.6425	4.4371
	0.8176	0.7931	0.9031	0.9280	0.9315	0.9501
Average	28.3294	28.1676	32.76	33.1037	33.2993	34.1354
	6.1055	6.0852	3.9510	3.6508	3.6797	3.4205
	0.8517	0.8442	0.9206	0.9324	0.9229	0.9564