Lightweight underwater image enhancement network based on GAN

Model	PSNR（G（X），Y）		SSIM（G（X），Y）		UIQM（G（X），Y）
Model	EUVP Input：27.623 $\pm$ 2.88	UFO-120 Input：26.98 $\pm$ 2.85	EUVP Input：0.784 $\pm$ 0.06	UFO-120 Input：0.773 $\pm$ 0.04	EUVP Input：2.458 $\pm$ 0.52	UFO-120 Input：2.755 $\pm$ 0.60
Res-GAN	24.526 $\pm$ 2.23	23.379 $\pm$ 2.35	0.692 $\pm$ 0.05	0.648 $\pm$ 0.07	2.585 $\pm$ 0.37	2.653 $\pm$ 0.55
UGAN-P	26.614 $\pm$ 2.77	27.121 $\pm$ 2.54	0.811 $\pm$ 0.06	0.785 $\pm$ 0.05	2.793 $\pm$ 0.39	2.713 $\pm$ 0.50
Pix2Pix	26.847 $\pm$ 2.97	26.777 $\pm$ 2.69	0.809 $\pm$ 0.05	0.785 $\pm$ 0.06	2.765 $\pm$ 0.42	2.742 $\pm$ 0.49
Cycle-GAN	28.869 $\pm$ 3.60	25.666 $\pm$ 3.70	0.894 $\pm$ 0.04	0.785 $\pm$ 0.60	2.949 $\pm$ 0.42	2.763 $\pm$ 0.57
FUnIE-GAN	26.325 $\pm$ 3.07	26.290 $\pm$ 2.76	0.791 $\pm$ 0.06	0.766 $\pm$ 0.06	2.839 $\pm$ 0.39	2.723 $\pm$ 0.49
FUnIE-GAN-up	25.905 $\pm$ 2.66	—	0.790 $\pm$ 0.06	—	2.809 $\pm$ 0.45	—
DUnet-GAN	26.047 $\pm$ 3.24	26.481 $\pm$ 3.02	0.798 $\pm$ 0.59	0.787 $\pm$ 0.06	2.848 $\pm$ 0.42	2.733 $\pm$ 0.51
DUnet-GAN-up	26.016 $\pm$ 2.74	—	0.792 $\pm$ 0.05	—	2.822 $\pm$ 0.41	—

Tools

Get Citation

Copy Citation Text

Hao-xuan LIU, Shan-ling LIN, Zhi-xian LIN, Tai-liang GUO, Jian-pu LIN. Lightweight underwater image enhancement network based on GAN[J]. Chinese Journal of Liquid Crystals and Displays, 2023, 38(3): 378

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Research Articles

Received: Jun. 24, 2022

Accepted: --

Published Online: Apr. 3, 2023

The Author Email: Jian-pu LIN (ljp@fzu.edu.cn)

DOI:10.37188/CJLCD.2022-0212

Topics