基于改进CycleGAN的多失真类型水下图像增强

吕振鸣,董绍江,夏宗佑,牟小燕,王明权

Multi-distortion type underwater image enhancement based on improved CycleGAN

Zhenming LV,Shaojiang DONG,Zongyou XIA,Xiaoyan MOU,Mingquan WANG

表 4 几种算法在EUVP和UIEB中彩色失真图像上的定量评估结果

Tab.4 Quantitative evaluation results of color distorted images on EUVP and UIEB using several algorithms

算法 PSNR SSIM UIQM UCIQE CLAHE 14.7715 0.6505 2.6781 0.3885 UDCP 16.0021 0.7073 2.8102 0.3936 Retinex 13.6909 0.6089 2.5434 0.3232 IBLA 20.6896 0.8120 3.2844 0.4105 CycleGAN 19.8091 0.7745 3.3273 0.4468 FUnIE-GAN 17.2722 0.7916 3.3072 0.4263 LANet 24.5578 0.8252 3.5305 0.4524 SCEIR 24.7437 0.8340 3.4496 0.4639 MGCycleGAN 25.0056 0.8397 3.5931 0.4676