基于特征细化与注意力增强重构的水下图像增强算法

万刚,王小波,石纲,叶德震,朱思思,司帆

Underwater image enhancement algorithm based on feature refinement and attention-augmented reconstruction

Gang WAN,Xiaobo WANG,Gang SHI,Dezhen YE,Sisi ZHU,Fan SI

表 2 各算法在UIEB数据集中的图像质量客观评价指标

Tab.2 Objective evaluation metrics of image quality for various algorithms on UIEB

算法 MSE↓ PSNR↑ SSIM↑ Entropy UIQM↑ UCIQE↑ 原始图像 1751.358 17.253 0.754 6.894 3.967 0.539 HE[7] 1868.491 16.621 0.737 7.167 4.294 0.664 CLAHE[8] 1216.226 18.194 0.713 7.319 4.424 0.567 DEFOG[6] 2730.896 15.243 0.709 6.841 4.231 0.553 Ucolor[25] 868.585 20.151 0.748 7.368 4.211 0.608 URST[26] 864.725 20.178 0.769 7.415 4.359 0.584 U-Shape[18] 734.626 20.881 0.709 7.262 4.397 0.578 WFFP[27] 828.485 18.949 0.791 7.749 4.213 0.609 Spectroformer[28] 1546.455 16.235 0.684 6.552 3.723 0.521 本研究算法 438.480 21.711 0.792 7.301 4.443 0.579