Skip to main content

Table 1 Performance evaluation of FA image generation methods

From: Synthesizing multi-frame high-resolution fluorescein angiography images from retinal fundus images using generative adversarial networks

 

HrGAN

Hervella’s method

Star GAN

Li’s method

Bicycle GAN

VtGAN

Unet

Our method

SSIM

0.6924

0.7081

0.6013

0.7194

0.6781

0.7034

0.6840

0.7126

PSNR(dB)

15.67

15.50

12.50

15.63

13.71

15.60

15.63

15.77

NCC

0.7421

0.6573

0.5621

0.7393

0.6901

0.7085

0.7030

0.7699

  1. The bold data represent the best value obtained