针对深海载人潜水器人造光源照射造成深海图像颜色失真、散射模糊以及光照不均匀等图像降质问题，提出一种基于多约束先验的贝叶斯Retinex深海载人潜水器图像增强方法。该方法首先通过一种基于统计的颜色校正方法对图像进行色彩校正处理，然后对光照图先后进行平滑、结构和不均匀光照高亮区域3种先验，并将3种先验条件结合到贝叶斯模型当中，并选取Lab颜色空间中L分量作为初始光照图来进行光照图估计的最优化求解。最后，通过伽马校正的方法对光照图和反射图进行处理，获得增强后的深海图像。实验结果表明，所提出的方法平均运行时间4.39 s，具有较低的复杂度，更加适合深海恶劣环境下的图像增强，处理得到的深海图像具有更好的观测效果。
A Bayesian Retinex image enhancement method based on multi-constraint prior was proposed to solve the problem of image degradation caused by artificial light source illumination of deep-sea Human Occupied Vehicle, such as color distortion, scattering blur and uneven illumination. The method first through a statistical color correction method based on the image color correction processing, and then the illumination map successively smoothing, structure and uneven illumination highlight region three prior, and the three prior conditions into the Bayesian model, and select the L component in the Lab color space as the initial illumination map to optimize the illumination map estimation. Finally, enhanced deep sea images were obtained by gamma correction for the illumination and reflection maps. The experimental results show that the proposed method has an average running time of 4.39 s, which has low complexity and is more suitable for image enhancement work in harsh deep-sea environments, and the processed deep-sea images have better observation effect.
2024,46(2): 143-149 收稿日期：2022-12-07
DOI：10.3404/j.issn.1672-7649.2024.02.025
分类号：TP391.9
基金项目：国家重点研发计划项目（2016YFC0300604）；辽宁省教育厅高等学校基本科研项目（重点项目）（LJKZ0442）
作者简介：秦豪(1995-),男,硕士研究生,研究方向为水下图像增强、图像处理、水下目标跟踪
参考文献：
[1] GUO Y, LI H, ZHUANG P. Underwater image enhancement using a multiscale dense generative adversarial network[J]. IEEE Journal of Oceanic Engineering, 2022, 45(3): 862-870.
[2] ZHUANG P, LI C, WU J. Bayesian retinex underwater image enhancement[J]. Engineering Applications of Artificial Intelligence, 2021, 101(1): 104171.
[3] ZHANG W, LI G, YING Z. A New underwater image enhancing method via color correction and illumination adjustment[C]// 2017 IEEE Visual Communications and Image Processing, St Petersburg, FL, 2017.
[4] 刘卫东, 李吉玉, 张文博, 等. 基于Retinex和ADMM优化的水下光照不均匀图像增强算法[J]. 西北工业大学学报, 2021, 39(4): 824-830.
[5] FU X , FAN Z, MEI L, et al. Two-step approach for single underwater image enhancement[C]// 2017 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS). 2017.
[6] 王丹, 张子玉, 赵金宝, 等. 基于场景深度估计的自然光照水下图像增强方法[J]. 机器人, 2021, 43(3): 364-372.
WANG D, ZHANG Z Y, ZHAO J B, et al. An enhancement method for underwater images under natural illumination based on scene depth estimation[J]. Robot, 2021, 43(3): 364-372.
[7] ZHUANG P, DING X. Underwater image enhancement using an edge-preserving filtering Retinex algorithm[J]. Multimedia Tools and Applications, 2020, 79(1): 1-21.
[8] FU X, ZHUANG P, HUANG Y, et al. A retinex-based enhancing approach for single underwater image[J]. 2014 IEEE International Conference on Image Processing. 2014: 4572–4576.
[9] TANG Zhijie, JIANG Lizhou, LUO Zhihang. A new underwater image enhancement algorithm based on adaptive feedback and Retinex algorithm[J]. Multimedia Tools and Applications, 2021.
[10] GUO X, YU L, LING H. LIME: Low-light image enhancement via illumination map estimation[J]. IEEE Transactions on Image Processing, 2017, 26(2): 982-993.
[11] KONG X, LIU L, QIAN Y. Low-Light image enhancement via poisson noise aware retinex model[C]// IEEE Signal Processing Letters, 2017.
[12] 郭彩云. 光照不均匀图像的增强方法研究[D]. 北京: 华北电力大学, 2020.
[13] 王梓丞, 尹勇. 基于色彩平衡及校正的水下图像增强算法[J]. 舰船科学技术, 2021, 43(21): 154-159.
[14] LI C, GUO J, CHEN S, et al. Underwater image restoration based on minimum information loss principle and optical properties of underwater imaging[J]. IEEE International Conference on Image Processing. (ICIP), IEEE, 2016, 993–1997.
