基于水体衰减系数反演的水下图像复原方法

doi:10.19678/j.issn.1000-3428.0064328

摘要/Abstract

摘要： 光在水中传输时容易受到水体和水中微粒的吸收和散射作用，造成成像模糊和颜色失真，导致水下图像质量有待提高。提出一种基于水体衰减系数反演的图像复原新方法，不预设水体类型和波长值，而是通过图像反演求得最符合图像特征的衰减系数。对图像去除垂直衰减色差，使描述水下图像强度规律的先验信息能更准确地反映水平方向的光传输特性，提升背景光估计和场景深度估计的准确性。利用水下图像的雾化模糊特征、红通道分量衰减特征以及通道间衰减差异特征得到水下图像场景深度的融合估计结果，依据像素场景深度与水下成像模型，通过非线性拟合直接反演出水下图像的衰减系数，并基于灰度世界假设对衰减系数进行优化，以完成传输率求解，实现图像复原。最后，使用非锐化掩膜增强图像，凸显细节信息，提升视觉效果。实验结果表明，与SMBOT、IBLA、ULAP、UDCP方法相比，该方法在RUIE数据集上的复原图像信息熵平均值分别提升了4.9%、5.4%、9.2%、17.9%，能较好地提升图像整体视觉效果。

关键词: 水下图像复原, 深度估计, 衰减系数, 非线性拟合, 细节增强

Abstract: When light is transmitted in water, it is easily absorbed and scattered by water bodies and water particles; this leads to image blurring and color distortion, which results in underwater image quality being improved.This study proposes a new image restoration method based on the inversion of the water attenuation coefficient.Instead of presetting the water type and wavelength value, the attenuation coefficient that best conforms to the image characteristics is obtained via image inversion.First, the global vertical chromatic aberration of the underwater image is eliminated to ensure that the prior information based on color attenuation can accurately reflect the transmission characteristics of light in the horizontal direction, then the global background light of the underwater image is estimated using the intensity characteristics of red and blue-green channels in the background region.Second, using the feature of blurriness, the red channel component attenuation, and the light attenuation difference, the fusion estimate of scene depth is obtained. According to the underwater imaging model and the scene depth of pixels, the attenuation coefficient of the underwater image is directly inversed via nonlinear fitting, then the attenuation coefficient is optimized based on Gray World assumption, the transmission rate is obtained, and the image is restored.Finally, the unsharp masking is used to further enhance the image to highlight the details and improve the visual effect.Experiments show that compared with the SMBOT, IBLA, ULAP, and UDCP methods, the average information entropy of the restored image on the RUIE dataset of this method is improved by 4.9%, 5.4%, 9.2%, and 17.9%, respectively, thereby further improving the overall visual effect of the image.

Key words: underwater image restoration, depth estimation, attenuation coefficient, nonlinear fitting, details enhancing

中图分类号:

TP391.41

庄子杰, 范之国, 金海红, 宫凯强. 基于水体衰减系数反演的水下图像复原方法[J]. 计算机工程, 2023, 49(1): 258-269.

ZHUANG Zijie, FAN Zhiguo, JIN Haihong, GONG Kaiqiang. Underwater Image Restoration Method Based on Water Attenuation Coefficient Inversion[J]. Computer Engineering, 2023, 49(1): 258-269.

http://www.ecice06.com/CN/Y2023/V49/I1/258

图/表 17

20230701181343

20230701181346

20230701181350

20230701181353

20230701181356

20230701181359

20230701181403

20230701181406

20230701181409

20230701181412

20230701181415

20230701181420

20230701181423

20230701181426

20230701181430

20230701181433

20230701181437

参考文献

[1] ANCUTI C O, ANCUTI C.Single image dehazing by multi-scale fusion[J].IEEE Transactions on Image Processing, 2013, 22(8):3271-3282.
[2] 唐永鹤, 蒋烈辉, 侯一凡, 等.基于Retinex与STFT的非接触指纹图像增强[J].计算机工程, 2017, 43(2):248-251, 256. TANG Y H, JIANG L H, HOU Y F, et al.Contactless fingerprint image enhancement based on Retinex and STFT[J].Computer Engineering, 2017, 43(2):248-251, 256.(in Chinese)
[3] 黄冬梅, 王龑, 宋巍, 等.不同颜色模型下自适应直方图拉伸的水下图像增强[J].中国图象图形学报, 2018, 23(5):640-651. HUANG D M, WANG Y, SONG W, et al.Underwater image enhancement method using adaptive histogram stretching in different color models[J].Journal of Image and Graphics, 2018, 23(5):640-651.(in Chinese)
[4] 李莉, 王新强, 银珊.基于衰减补偿与直方图拉伸的水下图像增强算法[J].计算机工程, 2022, 48(6):222-227. LI L, WANG X Q, YIN S.Underwater image enhancement algorithm based on attenuation compensation and histogram stretching[J].Computer Engineering, 2022, 48(6):222-227.(in Chinese)
[5] NARASIMHAN S G, NAYAR S K.Vision and the atmosphere[J].International Journal of Computer Vision, 2002, 48(3):233-254.
[6] 王红茹, 张弓, 卢道华, 等.基于背景光估计与颜色修正的水下图像增强[J].计算机工程, 2020, 46(10):253-258. WANG H R, ZHANG G, LU D H, et al.Underwater image enhancement based on background light estimation and color correction[J].Computer Engineering, 2020, 46(10):253-258.(in Chinese)
[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.
[8] DREWS JR P, DO NASCIMENTO E, MORAES F, et al.Transmission estimation in underwater single images[C]//Proceedings of IEEE International Conference on Computer Vision Workshops.Washington D.C., USA:IEEE Press, 2013:825-830.
[9] SONG W, WANG Y, HUANG D M, et al.Enhancement of underwater images with statistical model of background light and optimization of transmission map[J].IEEE Transactions on Broadcasting, 2020, 66(1):153-169.
[10] GALDRAN A, PARDO D, PICON A, et al.Automatic red-channel underwater image restoration[J].Journal of Visual Communication and Image Representation, 2015, 26:132-145.
[11] SONG W, WANG Y, HUANG D M, et al.A rapid scene depth estimation model based on underwater light attenuation prior for underwater image restoration[M]//Advances in Multimedia Information Processing-PCM 2018.Berlin, Germany:Springer, 2018:678-688.
[12] PENG Y T, COSMAN P C.Underwater image restoration based on image blurriness and light absorption[J].IEEE Transactions on Image Processing, 2017, 26(4):1579-1594.
[13] MENAKER D, TREIBITZ T, AVIDAN S.Color restoration of underwater images[C]//Proceedings of the British Machine Vision Conference.London, British:Machine Vision Association, 2017:45-54.
[14] JERLOV N G.Marine optics[EB/OL].[2022-01-10].https://www.doc88.com/p-6691253051491.html.
[15] PREISENDORFER R W.Hydrologic optics[M].Honolulu, USA:National Technical Information Service, 1976.
[16] MOBLEY C D.Light and water:radiative transfer in natural waters[EB/OL].[2022-01-10].https://www.doc88.com/p-1886395587527.html.
[17] ZHAO X W, JIN T, QU S.Deriving inherent optical properties from background color and underwater image enhancement[J].Ocean Engineering, 2015, 94:163-172.
[18] LI C Y, GUO J C, CONG R 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.
[19] AKKAYNAK D, TREIBITZ T.A revised underwater image formation model[C]//Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition.Washington D.C., USA:IEEE Press, 2018:6723-6732.
[20] HE K M, SUN J, TANG X O.Guided image filtering[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(6):1397-1409.
[21] SOLONENKO M G, MOBLEY C D.Inherent optical properties of Jerlov water types[J].Applied Optics, 2015, 54(17):5392-5401.
[22] BUCHSBAUM G.A spatial processor model for object colour perception[J].Journal of the Franklin Institute, 1980, 310(1):1-26.
[23] LI C Y, GUO C L, REN W Q, et al.An underwater image enhancement benchmark dataset and beyond[J].IEEE Transactions on Image Processing, 2020, 29:4376-4389.
[24] LIU R S, FAN X, ZHU M, et al.Real-world underwater enhancement:challenges, benchmarks, and solutions under natural light[J].IEEE Transactions on Circuits and Systems for Video Technology, 2020, 30(12):4861-4875.
[25] YANG M, SOWMYA A.An underwater color image quality evaluation metric[J].IEEE Transactions on Image Processing, 2015, 24(12):6062-6071.

选择文件类型/文献管理软件名称

选择包含的内容