Journal of Computer Science and Technology

   

Real-time Underwater Image Enhancement using Adaptive Full-Scale Retinex

Xing-Gui Xu1(徐兴贵), Xiang-Suo Fan 2,*(樊香所) and Yong-Li Liu3(刘永利)   

  1. 1School of Information, Yunnan University of Finance and Economics, Kunming 650221, China
    2School of Electrical Electronics and Computer Science, Guangxi University of Science and Technology, Liuzhou 545006, China
    3Department of Information and Communication, College of the Chinese People's Armed Police Force, Chengdu 610213, China
  • Received:2020-10-30 Revised:2022-08-23 Accepted:2022-12-13
  • Contact: Xiang-Suo Fan E-mail:100002085@gxust.edu.cn
  • About author:Xiang-Suo Fan is an assistant professor in School of Electrical Electronics and Computer Science, Guangxi University of Science and Technology. He received his Ph.D degree in in-formation processing from University of Elec-tronic Science and Technology of China, Cheng-du, in 2019, and his B.S. degree in electronic en-gineering from Beijing Institute of Technology, Beijing, in 2015. Currently he is mainly interest-ed in image processing and objects detection. He has published over twenty papers in referred conferences and journals.

Current Retinex-based image enhancement methods with fixed scale filters cannot adapt to situations involving various depths of field and illuminations. In this paper, a simple but effective method based on adaptive full-scale Retinex (AFSR) is proposed to clarify underwater images or videos. First, we de-sign an adaptive full-scale filter that is guided by the optical transmission rate to estimate illumination components. Then, to reduce the computational complexity, we develop a quantitative mapping method instead of non-linear log functions for directly calculating the reflection component. Moreover, the proposed method is capable of real-time processing of underwater video using temporal coherence and Fourier transformations. Compared with eight state-of-the-art clarification methods, our method yields comparable or better results for image contrast enhancement, color-cast correction and clarity.


中文摘要

1、研究背景(context)
近年来,水下成像系统已广泛应用于海洋能源勘探、水下环境监测、军事应用等领域。然而,由于水对光的选择性吸收和光在透射介质中的散射,光学系统获取到的水下图像会发生颜色畸变、对比度降低等多种退化降质现象。值得一提的是,水下图像增强场景下难以获取真正的清晰参考图像,这使得如端到端的深度学习方法将难以用于该类图像的增强处理。因此,针对水下图像进行有效的增强处理是一项重要而富有挑战性的任务。
2、目的(Objective)
针对水下降质图像特性,研究无需高清参考图像便能有效实现图像颜色和对比度增强的处理方法。同时,无需手动设置参数,要降低方法的计算复杂度以满足实际应用中的实时性要求。
3、方法(Method)
提出了一种自适应全尺度Retinex(AFSR)方法来增强水下退化图像或视频。与传统使用一个或多个固定尺度进行光照估计的方法不同,我们采用了由光传输引导的全尺度设计。因此,估计的照明分量克服了手动设计比例参数的限制。此外,该方法利用了颜色恒常性,直接进行颜色校正,无需任何先验知识和后处理。同时,采用一种简单的基于线性逼近的策略代替对数函数,大大降低了算法的计算复杂度。
4、结果(Result & Findings)
对真实水下退化数据的增强实验结果表明,该方法能同时校正颜色投射和提高对比度,并且能够实现视频序列图像的实时处理(0.021s/帧)。该方法已与8种最新的算法进行了比较,实验结果表明:该方法在对比度增强和颜色转换方面都比其他8种方法具有更好的性能。
5、结论(Conclusions)
本文中,我们提出了一种新的自适应全尺度Retinex(AFSR)水下图像增强方法,该方法同时解决了颜色偏差和对比度降低的问题。同时,所提方法适应于在不同光照和深度的场景,扩展了传统多尺度Retinex方法依赖于手动设计尺度参数的局限。与其他“级联”恢复方法(校正颜色和对比度问题的单个步骤)不同,该方法无需进行后处理便可以同时校正颜色和对比度。此外,本文提出的用线性映射函数代替对数运算直接计算反射分量的定量方法可以有效地降低计算复杂度,能够实现视频的实时增强处理。

Key words: underwater; image enhancement; Retinex; imaging through turbulent media;

[1] Zhi-Feng Xie, Shi Tang, Dong-Jin Huang, You-Dong Ding, Li-Zhuang Ma. Photographic Appearance Enhancement via Detail-based Dictionary Learning [J]. , 2017, 32(3): 417-429.
[2] Yan-Li Liu, Xiao-Gang Xu, Yan-Wen Guo, Jin Wang, Xin Duan, Xi Chen, and Qun-Sheng Peng, Senior Member, CCF. Pores-Preserving Face Cleaning Based on Improved Empirical Mode Decomposition [J]. , 2009, 24(3): 557-567.
[3] Yong-Ji Wang,Matthew Cartmell,Qiu-Ming Tao,and and Han Liu. A Generalized Real-Time Obstacle Avoidance Method Without the Cspace Calculation [J]. , 2005, 20(6): 774-787 .
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!

ISSN 1000-9000(Print)

         1860-4749(Online)
CN 11-2296/TP

Home
Editorial Board
Author Guidelines
Subscription
Journal of Computer Science and Technology
Institute of Computing Technology, Chinese Academy of Sciences
P.O. Box 2704, Beijing 100190 P.R. China
Tel.:86-10-62610746
E-mail: jcst@ict.ac.cn
 
  Copyright ©2015 JCST, All Rights Reserved