针对水下图像通常受到色偏、噪声和细节模糊的影响，且难以适应于各类视觉任务的问题，本文提出一种基于色彩衰减补偿和改进的多尺度Retinex的水下图像增强方法。首先针对水下图像出现色偏的问题，将水下图像分离为强、中、弱3个通道，且分别对3个通道进行不同尺度的颜色补偿，其次通过改进的多尺度Retinex算法增加边缘细节和对比度，在估计图像的光照分量时，用引导滤波替换高斯滤波，最后结合自适应伽马函数对光照分量进行校正。实验结果表明，本文方法能够有效改善不同场景下水下图像出现的色偏问题，能够提高对比度的同时增加图像细节，让水下图像的质量得到明显提升。
Aiming at the problem that underwater images are usually affected by color deviation, noise and detail blurring, and it is difficult to be adapted to all kinds of visual tasks, this paper proposes a method of underwater image enhancement based on color attenuation compensation and improved multi-scale Retinex. Firstly, to address the problem of color deviation in underwater images, the underwater images are separated into three channels: strong, medium and weak, and different scales of color compensation are applied to the three channels, secondly, the edge details and contrast are increased by the improved multi-scale Retinex algorithm, and the Gaussian filter is replaced by the guided filter in estimating the light component of the images, and finally the light component is corrected by combining with the adaptive gamma function. The experimental results show that the method in this paper can effectively improve the color bias problem occurring in underwater images under different scenes, and can improve the contrast while increasing the image details, so that the quality of underwater images can be significantly improved.
2024,46(18): 143-149 收稿日期：2023-11-9
DOI：10.3404/j.issn.1672-7649.2024.18.025
分类号：TP751.1
基金项目：国家自然科学基金资助项目（61702234）；船舶总体性能创新研究开放基金项目（25422217）
作者简介：胡记文(2000-),男,硕士研究生,研究方向为计算机视觉、图像处理
参考文献：
[1] LI Chongyi, QU jichang, PANG Yanwei. Single underwater image restoration by blue-green channels dehazing and red channel correction[C]//2016 IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP), 2016: 1371-1375.
[2] MISHRA A, GUPTA M, SHARMA P. Enhancement of underwater images using improved CLAHE[C]//International Conference on Advanced Computation and Telecommunication(ICACAT), 2018: 1-6.
[3] HE K M, SUN J, TANG X O. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(12): 2341-2353.
[4] DREWS J P, NASCIMENTO E, MORAES F, et al. Transmission estimation in underwater single images[C]//IEEE International Conference on Computer Vision Workshops, 2013: 825-83.
[5] ANCUTI C O, ANCUTI C, DEVLEESCHOUWER C, et al. Color balance and fusion for underwater image enhancement[J]. IEEE Transactions on Image Processing, 2018, 27(1): 379-393.
[6] FU X, ZHUANG P, HUANG Y, et al. A retinex-based enhancing approach for single underwater image[C]//IEEE international conference on image processing(ICIP), 2014: 4572–4576.
[7] REN W, LIU S, MA L, et al. Low-light image enhancement via a deep hybrid network[J]. IEEE Transactions on Image Process, 2019, 28(9): 4364-4375.
[8] LI C, GUO J, GUO C. Emerging from water: underwater image color correction based on weakly supervised color transfer[J]. IEEE Signal Processing Letters, 2018, 25(3): 323–327.
[9] LI J, SKINNER KA, EUSTICE RM, et al. WaterGAN: unsupervised generative network to enable real-time color correction of monocular underwater images[J]. IEEE Robot Autom letters, 2017, 3(1): 387–394.
[10] FABBRI C, ISLAM MJ, SATTAR J. Enhancing underwater imagery using generative adversarial net-works[C]//IEEE International Conference on Robotics and Automation(ICRA), 2018: 7159–7165.
[11] GUO Y, LI H, ZHUANG P. Underwater image enhancement using a multiscale dense generative adversarial network[J]. IEEE Journal of Ocean Engineering, 2019, 45(3): 862-870.
[12] LI C, GUO C, REN W, et al. An underwater image enhancement benchmark dataset and beyond[J]. IEEE Trans Image Process, 2020, 29(10): 4376-4389.
[13] 田宁, 程莉, 元海文, 等. 基于Retinex模型的水下图像增强方法[J]. 中国科技论文, 2022, 17(11): 1281-1288.
[14] HO S L, SANG W M, LL K E. Underwater image enhancement using successive color correction and superpixel dark channel prior[J]. Symmetry, 2020, 12(8): 3-5.
[15] 武凌霄, 康家银, 姬云翔. NSST域下基于引导滤波与稀疏表示的红外与可见光图像融合[J]. 红外技术, 2023, 45(9): 915-924.
[16] WANG M, ZHENG S, LI X, et al. A new image denoising method based on Gaussian filter[C]//IEEE International Conference on Information Science, Electronics and Electrical Engineering, 2014: 163-167.
[17] 刘志成, 王殿伟, 刘颖, 等. 基于二维伽马函数的光照不均匀图像自适应校正算法[J]. 北京理工大学学报, 2016, 36(02): 191-196+214.
[18] 贾芃, 李博, 赵晓龙. 基于HSI的改进Retinex水下图像增强算法[J]. 实验室研究与探索, 2020, 39(12): 1-4+14.
[19] PANETTA K, GAO C, AGAIAN S. Human-visual-system-inspired underwater image quality measures[J]. IEEE Journal Oceanic Engineering, 2016, 41(3): 541–551.
[20] YANG Miao, ARCOT Sowmya. An underwater color image quality evaluation metric[J]. IEEE transactions on image process, 2015, 21(12): 6062–6071.
