留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

海底主动光学探测影像亮度校正与色彩恢复

刘镕滔 柳稼航

刘镕滔, 柳稼航. 海底主动光学探测影像亮度校正与色彩恢复[J]. 中国光学(中英文), 2022, 15(4): 689-702. doi: 10.37188/CO.2021-0211
引用本文: 刘镕滔, 柳稼航. 海底主动光学探测影像亮度校正与色彩恢复[J]. 中国光学(中英文), 2022, 15(4): 689-702. doi: 10.37188/CO.2021-0211
LIU Rong-tao, LIU Jia-hang. Brightness correction and color restoration of seabed image obtained by active optical detection[J]. Chinese Optics, 2022, 15(4): 689-702. doi: 10.37188/CO.2021-0211
Citation: LIU Rong-tao, LIU Jia-hang. Brightness correction and color restoration of seabed image obtained by active optical detection[J]. Chinese Optics, 2022, 15(4): 689-702. doi: 10.37188/CO.2021-0211

海底主动光学探测影像亮度校正与色彩恢复

doi: 10.37188/CO.2021-0211
基金项目: 高分辨率对地观测系统计划(No. 41-Y30F07-9001-20/22);江苏省SC人才项目(No. JSSCRC2021501)
详细信息
    作者简介:

    刘镕滔(1997—),男,湖南娄底人,2019年于重庆大学通信工程专业获得学士学位,现于南京航空航天大学航天学院攻读通信与信息系统专业硕士研究生,研究方向为图像复原与增强。E-mail:lrt@nuaa.edu.cn

    柳稼航(1977—),男,博士,教授,2000年于武汉大学获得摄影测量与遥感工学学士学位;2003年于中国地震局地质研究所获得构造地质理学硕士学位;2011年于上海交通大学获得模式识别与智能系统工学博士学位。2003年—2019年,在中国科学院西安光学精密机械研究所工作,历任实习研究员,助理研究员、副研究员、研究员。现为南京航空航天大学航天学院教授,博士生导师。主要从事遥感技术,图像处理,机器视觉及人工智能等方面的研究。E-mail:jhliu@nuaa.edu.cn

  • 中图分类号: TP391

Brightness correction and color restoration of seabed image obtained by active optical detection

Funds: Supported by the China High Resolution Earth Observation System Program (No. 41-Y30F07-9001-20/22); Innovative talent program of Jiangsu (No. JSSCRC2021501)
More Information
  • 摘要:

    主动光学成像探测是海底形貌与环境探测的重要方式,广泛应用于大洋勘探、海底探测等领域。然而,由于海水对光的衰减作用,造成光学影像照度不均、颜色失真、对比度低等质量退化问题。本文依据水下主动光学成像探测的特点,提出了一种基于相对辐射校正原理的水下图像增强方法。该方法将增强过程分为亮度补偿和色彩恢复两个阶段。在亮度补偿阶段,依据水下点光源的成像特点和辐射衰减机制,采用相对辐射校正原理对水下图像分通道进行补偿,消除因光源不均、光程不同等因素造成的亮度畸变。在色彩恢复阶段,首先对红通道图像进行自适应补偿和色彩粗平衡,在此基础上进一步利用Retinex模型对图像进行色彩恢复。利用实际的海底勘探图像进行实验验证,结果表明本文方法的增强结果亮度均匀、色彩自然,有效提升了图像质量。相较现有方法,本文方法的结果无论主观感受还是客观评价整体更优。同时,由于本文方法不需要光源、相机等特性参数,仅利用实际观测图像本身进行校正,因而具有更好的适应性。

     

  • 图 1  点光源垂直照射下的照度相对分布示意图。(a)理想点光源垂直照射示意图;(b)光照强度衰减示意图

    Figure 1.  Relative distribution of illuminance under vertical illumination of the point light source . (a) Schematic of vertical illumination of the ideal point light source and (b) schematic of light intensity attenuation

    图 2  海底主动光学成像示意图

    Figure 2.  Schematic diagram of seabed active optical imaging

    图 3  点光源照明下的均匀海底光学图像

    Figure 3.  Underwater optical image of a uniform seafloor illuminated by the point light source

    图 4  参考图像的各通道图像及其亮度拟合曲面。(a)~(c)参考图像的红、绿、蓝通道;(d)~(f)红、绿、蓝通道的三维俯视图;(g)~(i)红、绿、蓝通道的拟合曲面

    Figure 4.  Each channel of the reference image and their brightness fitting surface (a)−(c) red, green, blue channels of the reference image; (d)−(f) 3D top view of the red, green and blue channels; (g)−(i) fitting surface of the red, green, blue channels

    图 5  参考图像分通道亮度补偿结果。(a)~(c)分别为红、绿、蓝三通道的亮度补偿量;(d)~ (f)补偿后的红、绿、蓝单通道影像

    Figure 5.  Individual brightness compensation for each channel of the reference image (a)−(c) Degree of compensation of the red, green and blue channels; (d)−(f) compensated red, green and blue single channel images

    图 6  采用参考图像校正参数对同批次的图像补偿。(a)~(c)红绿蓝三通道原始灰度图像;(d)~(f)补偿后的红绿蓝影像

    Figure 6.  The compensation process of the image in the same batch with reference image correction parameters. (a)−(c) Original gray images of the red, green and blue channels, (d)−(f) compensated images of the three channels

    图 8  不同算法对参考图像校正的结果。(a)MSR;(b)CLAHE;(c)RCP;(d)RWCGC;(e)UWCNN;(f)本文方法

    Figure 8.  Comparison of correction results of the underwater image processed by different algorithms. (a) MSR; (b) CLAHE; (c) RCP; (d) RWCGC; (e) UWCNN; (f) The proposed method

    图 9  水下图像常用增强处理算法的结果对比。(a)原始图片;(b)~(g)分别使用MSR、CLAHE、RCP、RWCGC、UWCNN和本文方法的增强结果

    Figure 9.  Comparison of underwater image enhancement processed by different algorithms. (a) Original images; (b)−(g) the results obtained by MSR, CLAHE, RCP, RWCGC, UWCNN and the proposed method

    图 7  不同图像增强方法效果对比。(a)预补偿输入图像;(b)~(f)分别是灰色世界、本文预处理方法、MSR、灰色世界算法+MSR、本文预处理方法+MSR方法结果

    Figure 7.  Comparison of image enhancement with different pre-processing methods. (a) Pre-compensation image; (b)−(f) compensated results with Gray world, the proposed pre-processing method, MSR, Gray world + MSR, and the proposed pre-processing method + MSR methods

    表  1  参考图像各通道灰度曲面四次多项式拟合参数

    Table  1.   Quartic polynomial fitting parameters of the gray surface of each channel of the reference image

    参数R通道G通道B通道
    a0019.2546.3250.18
    a10−0.10060.42080.4696
    a01−0.11520.39040.4299
    a201.005×10−3−4.915×10−4 −7.192×10−4
    a111.038×10−3−6.395×10−4−8.548×10−4
    a021.002×10−3−4.785×10−4−6.328×10−4
    a30−2.241×10−61.353×10−76.774×10−7
    a21−1.584×10−6−2.121×10−7−1.141×10−7
    a12−7.252×10−72.581×10−63.145×10−6
    a03−2.75×10−6−8.219×10−7−5.752×10−7
    a401.391×10−9−8.548×10−10−1.382×10−9
    a313.485×10−103.321×10−94.136×10−9
    a222.072×10−9−4.899×10−9−6.476×10−9
    a13−1.332×10−94.571×10−101.053×10−9
    a042.382×10−93.993×10−107.139×10−11
    下载: 导出CSV

    表  2  参考图像及其在不同算法处理后的客观评价结果

    Table  2.   Objective image quality evaluation results of the reference image and images processed by different algorithms

    指标参考图像MSRCLAHERCPRWCGCUWCNNOurs
    UIQM0.93223.8092.31933.35353.15081.42384.0546
    UCIQE20.398824.981325.617324.043328.784721.454628.2705
    注:加粗字体为每行最优值。
    下载: 导出CSV

    表  3  对于图9第一列影像不同算法的客观评价结果

    Table  3.   Objective image quality evaluation results of the first column of Fig. 9 processed by different algorithms

    指标图9第一列原图MSRCLAHERCPRWCGCUWCNNOurs
    UIQM1.01473.85572.99082.69693.64443.02114.5062
    UCIQE18.257528.071324.380222.439729.377821.901729.8468
    注:加粗字体为每行最优值。
    下载: 导出CSV

    表  4  对于图9第二列影像不同算法的客观质量评价指标

    Table  4.   Objective image quality evaluation indexes of the second column of Fig. 9 processed by different algorithms

    指标图9第二列原图MSRCLAHERCPRWCGCUWCNNOurs
    UIQM1.01473.85572.99082.69693.64443.02114.5583
    UCIQE20.556127.670927.543924.287729.077820.952329.5887
    注:加粗字体为每行最优值
    下载: 导出CSV

    表  5  对于图9第三列影像不同算法的客观质量评价指标

    Table  5.   Objective image quality evaluation indexes of the third column of Fig. 9 processed by different algorithms

    图9第三列原图MSRCLAHERCPRWCGCUWCNNOurs
    UIQM0.67393.71422.83592.46113.121462.61834.3095
    UCIQE20.031228.139225.583123.894329.099622.886829.1635
    注:加粗字体为每行最优值
    下载: 导出CSV
  • [1] 邓翔宇, 王惠刚, 张永庆. 基于主动光照的深海图像增强算法[J]. 光子学报,2020,49(3):0310001. doi: 10.3788/gzxb20204903.0310001

    DENG X Y, WANG H G, ZHANG Y Q. Deep sea image enhancement method based on the active illumination[J]. Acta Photonica Sinica, 2020, 49(3): 0310001. (in Chinese) doi: 10.3788/gzxb20204903.0310001
    [2] 郭继昌, 李重仪, 郭春乐, 等. 水下图像增强和复原方法研究进展[J]. 中国图象图形学报,2017,22(3):273-287.

    GUO J CH, LI CH Y, GUO CH L, et al. Research progress of underwater image enhancement and restoration methods[J]. Journal of Image and Graphics, 2017, 22(3): 273-287. (in Chinese)
    [3] HENKE B, VAHL M, ZHOU ZH L. Removing color cast of underwater images through non-constant color constancy hypothesis[C]. 2013 8th International Symposium on Image and Signal Processing and Analysis (ISPA), IEEE, 2013.
    [4] FU X Y, ZHUANG P X, HUANG Y, et al.. A retinex-based enhancing approach for single underwater image[C]. 2014 IEEE International Conference on Image Processing (ICIP), IEEE, 2014.
    [5] 杨卫中, 徐银丽, 乔曦, 等. 基于对比度受限直方图均衡化的水下海参图像增强方法[J]. 农业工程学报,2016,32(6):197-203. doi: 10.11975/j.issn.1002-6819.2016.06.027

    YANG W ZH, XU Y L, QIAO X, et al. Method for image intensification of underwater sea cucumber based on contrast-limited adaptive histogram equalization[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(6): 197-203. (in Chinese) doi: 10.11975/j.issn.1002-6819.2016.06.027
    [6] XIANG W D, YANG P, WANG SH, et al. Underwater image enhancement based on red channel weighted compensation and gamma correction model[J]. Opto-Electronic Advances, 2018, 1(10): 180024.
    [7] 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. doi: 10.1109/TPAMI.2010.168
    [8] CHIANG J Y, CHEN Y C. Underwater image enhancement by wavelength compensation and dehazing[J]. IEEE Transactions on Image Processing, 2012, 21(4): 1756-1769. doi: 10.1109/TIP.2011.2179666
    [9] GALDRAN A, PARDO D, PICÓN A, et al. Automatic red-channel underwater image restoration[J]. Journal of Visual Communication and Image Representation, 2015, 26: 132-145. doi: 10.1016/j.jvcir.2014.11.006
    [10] 谢昊伶, 彭国华, 王凡, 等. 基于背景光估计与暗通道先验的水下图像复原[J]. 光学学报,2018,38(1):0101002. doi: 10.3788/AOS201838.0101002

    XIE H L, PENG G H, WANG F, et al. Underwater image restoration based on background light estimation and dark channel prior[J]. Acta Optica Sinica, 2018, 38(1): 0101002. (in Chinese) doi: 10.3788/AOS201838.0101002
    [11] 王一斌, 尹诗白, 吕卓纹. 自适应背景光估计与非局部先验的水下图像复原[J]. 光学 精密工程,2019,27(2):499-510. doi: 10.3788/OPE.20192702.0499

    WANG Y B, YIN SH B, LÜ ZH W. Underwater image restoration with adaptive background light estimation and non-local prior[J]. Optics and Precision Engineering, 2019, 27(2): 499-510. (in Chinese) doi: 10.3788/OPE.20192702.0499
    [12] 林森, 刘世本, 唐延东. 多输入融合对抗网络的水下图像增强[J]. 红外与激光工程,2020,49(5):20200015. doi: 10.3788/irla.28_2020-0015

    LIN S, LIU SH B, TANG Y D. Multi-input fusion adversarial network for underwater image enhancement[J]. Infrared and Laser Engineering, 2020, 49(5): 20200015. (in Chinese) doi: 10.3788/irla.28_2020-0015
    [13] LI CH Y, ANWAR S, PORIKLI F. Underwater scene prior inspired deep underwater image and video enhancement[J]. Pattern Recognition, 2020, 98: 107038. doi: 10.1016/j.patcog.2019.107038
    [14] 刘群, 刘崇, 朱小磊, 等. 星载海洋激光雷达最佳工作波长分析[J]. 中国光学,2020,13(1):148-155. doi: 10.3788/co.20201301.0148

    LIU Q, LIU CH, ZHU X L, et al. Analysis of the optimal operating wavelength of spaceborne oceanic lidar[J]. Chinese Optics, 2020, 13(1): 148-155. (in Chinese) doi: 10.3788/co.20201301.0148
    [15] 全向前, 陈祥子, 全永前, 等. 深海光学照明与成像系统分析及进展[J]. 中国光学,2018,11(2):153-165. doi: 10.3788/co.20181102.0153

    QUAN X Q, CHEN X Z, QUAN Y Q, et al. Analysis and research progress of deep-sea optical illumination and imaging system[J]. Chinese Optics, 2018, 11(2): 153-165. (in Chinese) doi: 10.3788/co.20181102.0153
    [16] 吕宝林, 佟首峰, 徐伟, 等. 基于配准的机载红外非均匀性校正技术应用[J]. 中国光学,2020,13(5):1124-1137. doi: 10.37188/CO.2020-0109

    LV B L, TONG SH F, XU W, et al. Non-uniformity correction of airborne infrared detection system based on inter-frame registration[J]. Chinese Optics, 2020, 13(5): 1124-1137. (in Chinese) doi: 10.37188/CO.2020-0109
    [17] ANCUTI C O, ANCUTI C, DE VLEESCHOUWER C, et al. Color balance and fusion for underwater image enhancement[J]. IEEE Transactions on Image Processing, 2018, 27(1): 379-393. doi: 10.1109/TIP.2017.2759252
    [18] LAND E H, MCCANN J J. Lightness and retinex theory[J]. Journal of the Optical Society of America, 1971, 61(1): 1-11. doi: 10.1364/JOSA.61.000001
    [19] RAHMAN Z, JOBSON D J, WOODELL G A. Multi-scale retinex for color image enhancement[C]. Proceedings of 3rd IEEE International Conference on Image Processing, IEEE, 1996.
    [20] PANETTA K, GAO CH, AGAIAN S. Human-visual-system-inspired underwater image quality measures[J]. IEEE Journal of Oceanic Engineering, 2016, 41(3): 541-551. doi: 10.1109/JOE.2015.2469915
    [21] YANG M, SOWMYA A. An underwater color image quality evaluation metric[J]. IEEE Transactions on Image Processing, 2015, 24(12): 6062-6071. doi: 10.1109/TIP.2015.2491020
  • 加载中
图(9) / 表(5)
计量
  • 文章访问数:  612
  • HTML全文浏览量:  421
  • PDF下载量:  170
  • 被引次数: 0
出版历程
  • 收稿日期:  2021-12-06
  • 修回日期:  2022-01-10
  • 网络出版日期:  2022-05-16

目录

    /

    返回文章
    返回

    重要通知

    2024年2月16日科睿唯安通过Blog宣布,2024年将要发布的JCR2023中,229个自然科学和社会科学学科将SCI/SSCI和ESCI期刊一起进行排名!《中国光学(中英文)》作为ESCI期刊将与全球SCI期刊共同排名!