Application of Image Super-Resolution Reconstruction in Gymnastics Training by Using Internet of Things Technology

Table 1

PSNR performance comparison of traditional SISR and RGB image algorithms.

Dataset	Super-resolution scale	Bicubic	NE-LS	NE-LLE	ESC	A+
		PSNR (dB)	PSNR (dB)	PSNR (dB)	PSNR (dB)	PSNR (dB)
Set5	×2	32.34	34.34	34.45	34.46	35.22
Setl4	×2	29.03	30.51	30.58	30.62	31.09
Bl00	×2	28.01	29.05	29.09	29.08	29.45
Urban100	×2	25.59	27.21	27.18	27.17	27.93
D1V2K	×2	31.12	32.58	32.63	32.65	33.19
PIRM	×2	29.31	30.54	30.66	30.66	31.06
Set5	×3	29.09	30.47	30.54	30.60	31.28
Setl4	×3	26.35	27.41	27.42	27.49	27.95
Bl00	×3	25.85	26.53	26.55	26.58	26.88
Urban100	×3	23.18	24.17	24.13	24.17	24.76
D1V2K	×3	28.35	29.30	29.32	29.36	29.77
PIRM	×3	26.64	27.43	27.48	27.51	27.82
Set5	×4	27.13	28.27	28.33	28.41	29.00
Setl4	×4	24.81	25.63	25.63	25.70	26.13
Bl00	×4	24.63	25.16	25.18	25.22	25.47
Urban100	×4	21.86	22.59	22.56	22.62	23.06
D1V2K	×4	26.81	27.56	27.58	27.63	27.98
PIRM	×4	25.16	25.77	25.80	25.85	26.11