Skip to main content

Table 4 PSNR and SSIM after reconstruction of various murals with different algorithms

From: An improved algorithm for superresolution reconstruction of ancient murals with a generative adversarial network based on asymmetric pyramid modules

Image type

Index

BI

Literature [17]

Literature [20]

SRGAN

EDSR

ProGAN

Animals

PSNR/SSIM

20.92/0.44

27.36/0.63

27.81/0.64

28.05/0.66

28.44/0.68

28.56/0.68

Builds

PSNR/SSIM

19.89/0.44

24.21/0.57

25.15/0.62

25.41/0.63

25.62/0.64

25.85/0.65

Cloud

PSNR/SSIM

19.84/0.48

24.40/0.62

25.40/0.66

25.62/0.67

25.87/0.69

26.18/0.70

Disciple

PSNR/SSIM

20.21/0.46

25.88/0.62

26.70/0.62

26.81/0.66

27.15/0.68

27.35/0.69

Fo

PSNR/SSIM

20.06/0.45

24.05/0.56

24.88/0.58

25.06/0.61

25.23/0.62

25.49/0.63

People

PSNR/SSIM

20.23/0.43

26.92/0.64

27.50/0.63

27.75/0.67

28.12/0.69

28.28/0.70

Plant

PSNR/SSIM

21.20/0.50

26.75/0.66

27.44/0.65

27.60/0.70

28.05/0.72

28.21/0.72

Pusa

PSNR/SSIM

20.49/0.44

24.99/0.58

25.83/0.61

25.91/0.62

2 6.11/0.64

26.27/0.65