Underwater images from different scenes often exhibit complex and non-uniform degradation due to factors such as light absorption by water and the scattering effects of suspended particles. Even within the same image, the degradation degree varies across regions due to differences in scene depth. The most existing underwater image enhancement methods fail to specifically address the non-uniform degradation, leading to poor enhancement results. To solve this issue, this paper proposes an iterative underwater image enhancement network (IUIENet) based on degration distribution perception. IUIENet consists of three modules: a pre-enhancement module, a degradation distribution estimation module and an image refinement module. The pre-enhancement module initially estimates the enhancement result, while the degradation distribution estimation module and the image refinement module optimize the enhancement results using iterative cooperation. Experimental results demonstrate that IUIENet outperforms the compared methods in both visual quality and quantitative metrics on the UIEB, EUVP, and LSUI benchmark datasets.
迭代次数对结果的影响:为验证迭代次数对结果的影响,将IUIENet的迭代次数分别设置为0、1、2、3,具有不同迭代次数的模型分别记作IUIENet-0、IUIENet-1、IUIENet-2和IUIENet-3.当迭代次数为0时,表示仅使用预增强模块对结果进行估计.使用与IUIENet相同的损失函数与训练方式在UIEB数据集对两个变体进行训练,各个模型的性能如表4所示.可以看到,随着迭代次数的增加,模型的性能不断提升,但是提升的幅度在逐渐减小.其中,IUIENet-1的PSNR指标比IUIENet-0高出1.5 dB 以上,这充分说明了引入迭代机制的有效性.为了避免过多的训练成本,减少模型的推理时间,本文将迭代次数设置为2.
NARASIMHANS, NAYARS. Vision and the atmosphere[J]. International Journal of Computer Vision, 2002, 48: 233-254.
[2]
HUMMELR A. Histogram modification techniques[J]. Computer Graphics and Image Processing, 1975, 4(3): 209-224.
[3]
PIZERS M, AMBURNE P, AUSTINJ D,et al. Adaptive histogram equalization and its variations[J]. Computer Vision,Graphics,and Image Processing, 1987, 39(3): 355-368.
[4]
REZAA M. Realization of the contrast limited adaptive histogram equalization (CLAHE) for real-time image enhancement[J]. Journal of VLSI Signal Processing Systems for Signal,Image and Video Technology, 2004, 38(1): 35-44.
[5]
LIANGZ, DINGX Y, WANGY F,et al .GUDCP:generalization of underwater dark channel prior for underwater image restoration[J]. IEEE Transactions on Circuits and Systems for Video Technology,2022,32(7):4879-4884.
[6]
LIC Y, GUOJ C, CONGR M,et al .Underwater image enhancement by dehazing with minimum information loss and histogram distribution prior[J].IEEE Transactions on Image Processing,2016,25(12):5664-5677.
[7]
ZHUANGP X, WUJ M, PORIKLIF,et al .Underwater image enhancement with hyper-Laplacian reflectance priors[J].IEEE Transactions on Image Processing,2022,31:5442-5455.
[8]
WUJ J, LIUX L, LUQ H,et al .FW-GAN:Underwater image enhancement using generative adversarial network with multi-scale fusion[J].Signal Processing:Image Communication,2022,109:116855.
[9]
SUJ N, GANM, CHENG Y,et al .Global learnable attention for single image super-resolution[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2023,45(7):8453-8465.
[10]
ZHANGK K, ZHAOT S, CHENW L,et al .Perception-driven similarity-clarity tradeoff for image super-resolution quality assessment[J].IEEE Transactions on Circuits and Systems for Video Technology,2024,34(7): 5897-5907.
[11]
CHENW L, CAIB Q, ZHENGS M,et al .Perception-and-cognition-inspired quality assessment for sonar image super-resolution[J].IEEE Transactions on Multimedia,2024,26:6398-6410.
[12]
CHENG Y, WENGW D, SUJ N,et al .Dynamic degradation intensity estimation for adaptive blind super-resolution:a novel approach and benchmark dataset[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2024, 34(6):4762-4772.
[13]
SUJ N, GANM, CHENG Y, et al. High-similarity-pass attention for single image super-resolution[J]. IEEE Transactions on Image Processing, 2024, 33: 610-624.
YANGY, CHENY. End-to-end image dehazing algorithm based on joint mapping of two-branch features[J]. Journal of Hunan University (Natural Sciences), 2024,51(6):10-19.(in Chinese)
YANGY, WUX D, DUK .T-shaped image dehazing network based on wavelet transform and attention mechanism[J].Journal of Hunan University (Natural Sciences),2022,49(10):61-68.(in Chinese)
[20]
LIS S, ZHOUY, RENW Q,et al .PFONet:a progressive feedback optimization network for lightweight single image dehazing[J].IEEE Transactions on Image Processing,2023,32:6558-6569.
[21]
ZHUX S, LIS S, GANY D,et al .Multi-stream fusion network with generalized smooth L1 loss for single image dehazing[J].IEEE Transactions on Image Processing,2021,30:7620-7635.
[22]
ZHUZ J, YANGX G, LUR T,et al .Ghost imaging in the dark:a multi-illumination estimation network for low-light image enhancement[J].IEEE Transactions on Circuits and Systems for Video Technology, 2025, 35(2): 1576-1590.
[23]
SHEC Y, HANF J, WANGL D,et al .MPC-net:multi-prior collaborative network for low-light image enhancement[J].IEEE Transactions on Circuits and Systems for Video Technology,2024,34(10): 10385-10398.
WANGW Y, SHUC Y, ZHUL T,et al .Low-illumination image enhancement method based on Retinex and Gamma transformation[J].Journal of Hunan University (Natural Sciences),2024,51(10):136-144.(in Chinese)
[26]
ZHOUY, YANK M, LIX F .Underwater image enhancement via physical-feedback adversarial transfer learning[J]. IEEE Journal of Oceanic Engineering, 2022, 47(1): 76-87.
[27]
HUANGS R, WANGK Y, LIUH,et al .Contrastive semi-supervised learning for underwater image restoration via reliable bank[C]//2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver,BC,Canada. IEEE,2023:18145-18155.
[28]
ZHOUJ C, LIB S, ZHANGD H,et al .UGIF-net:an efficient fully guided information flow network for underwater image enhancement[J].IEEE Transactions on Geoscience and Remote Sensing,2023,61:4206117.
[29]
YANK, LIANGL Y, ZHENGZ Q,et al .Medium transmission map matters for learning to restore real-world underwater images[J].Applied Sciences,2022,12(11): 5420.
[30]
MUP, FANGJ, QIANH T,et al .Transmission and color-guided network for underwater image enhancement[C]//2023 IEEE International Conference on Multimedia and Expo (ICME). Brisbane,Australia. IEEE,2023:1337-1342.
[31]
WUS C, LUOT, JIANGG Y,et al .A two-stage underwater enhancement network based on structure decomposition and characteristics of underwater imaging[J]. IEEE Journal of Oceanic Engineering,2021, 46(4): 1213-1227.
[32]
CONGR M, YANGW Y, ZHANGW,et al .PUGAN:physical model-guided underwater image enhancement using GAN with dual-discriminators[J]. IEEE Transactions on Image Processing,2023, 32: 4472-4485.
[33]
LIC Y, ANWARS, HOUJ H,et al. Underwater image enhancement via medium transmission-guided multi-color space embedding[J]. IEEE Transactions on Image Processing,2021,30:4985-5000.
[34]
PENGL T, ZHUC L, BIANL H .U-shape transformer for underwater image enhancement[C]//Computer Vision-ECCV 2022 Workshops. Cham:Springer Nature Switzerland,2023:290-307.
[35]
SHARMAP, BISHTI,SUR A .Wavelength-based attributed deep neural network for underwater image restoration[J].ACM Transactions on Multimedia Computing,Communications,and Applications,2023,19(1):1-23.
[36]
KINGMAD P, BAJ, HAMMADM M .Adam:a method for stochastic optimization[EB/OL].
[37]
LIC Y, GUOC L, RENW Q,et al .An underwater image enhancement benchmark dataset and beyond[J]. IEEE Transactions on Image Processing, 2019, 29: 4376-4389.
[38]
ISLAMM J, XIAY Y, SATTARJ .Fast underwater image enhancement for improved visual perception[J].IEEE Robotics and Automation Letters,2020,5(2): 3227-3234.
[39]
DREWS JrP, NASCIMENTO EDO, MORAESF,et al .Transmission estimation in underwater single images[C]//2013 IEEE International Conference on Computer Vision Workshops. Sydney,NSW,Australia. IEEE, 2013: 825-830.
GUOC L, WUR Q, JINX, et al. Underwater ranker:learn which is better and how to be better[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2023,37(1):702-709.
[42]
KHANM R, MISHRAP, MEHTAN,et al .Spectroformer:multi-domain query cascaded transformer network for underwater image enhancement[C]//2024 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV). Waikoloa,HI,USA. IEEE,2024:1443-1452.
[43]
LINW T, LINY X, CHENJ W, et al. PixMamba: leveraging state space models in a dual-level architecture for underwater image enhancement[C]//Proceedings of the Asian Conference on Computer Vision. 2024: 3622-3637.