基于透射率融合与优化的水下图像复原
|
摘要
由于光与水体介质的相互作用的固有复杂性,所拍摄的水下场景图像发生降质退化,这导致诸如水下图像的低对比度以及色彩失真等问题.尤其是对于污染较为严重或者高浑浊度水体下获得的图像,典型的水下图像处理方法有一定的性能缺陷,而其根本原因在于其无法对于散射造成的色彩失真问题进行有效处理,同时也未充分考虑对比度信息带来的影响.因此,本文提出了一种透射率融合与优化方法,并由此给出有效的水下图像复原方案.该方案考虑各通道在水下衰减差异性,建立了基于颜色衰减差异的水下图像形成模型,其场景环境光由成像光源计算得到.首先,利用色彩校正算法来处理水下图像存在的色彩失真的问题,并利用灰度色调算法对场景光源的颜色进行估计;接着,基于水下场景约束,提出水体透射率估计方法,同时基于对比度先验提出对比度透射率估计算法;然后,将上述两种透射率进行融合,并使用多方向梯度加权正则化进行细化;最后,通过求解基于颜色衰减差异的水下图像形成模型复原图像.通过多次实验,所提出的基于透射率融合优化的水下图像复原方法得到的图像不仅具有相对自然的颜色,同时保持着良好的细节和对比度信息.
Images acquired in underwater environments undergo a degradation process due to the inherent complexity of the interaction of light with the medium. Traditional methods fail to restore underwater images with heavy turbidity.This is mainly because they cannot efficiently remove the color distortion caused by scattering,and the restored image is blurred because of lack of contrast information. Thus,this paper proposes a transmission fusion estimation method. First,considering the difference in attenuation of each channel,a underwater image formation model based on the difference in color attenuation is obtained by assuming the light source is the same as the ambient light. Then,an effective color correction approach is designed to remove the color distortion; meanwhile,estimate the light source by the gray level algorithm. Next,the transmission fusion estimation method is established based on the transmission characteristics of water and the contrast constraint prior and is refined with multi-direction gradient weighted regularization. Finally,the clear image is obtained by solve the new underwater image model. The experimental results shows the image restored by the proposed underwater image restoration method which based on transmission fusion and optimization not only has the relatively natural color,but also maintains better detail and contrast information.
Images acquired in underwater environments undergo a degradation process due to the inherent complexity of the interaction of light with the medium. Traditional methods fail to restore underwater images with heavy turbidity.This is mainly because they cannot efficiently remove the color distortion caused by scattering,and the restored image is blurred because of lack of contrast information. Thus,this paper proposes a transmission fusion estimation method. First,considering the difference in attenuation of each channel,a underwater image formation model based on the difference in color attenuation is obtained by assuming the light source is the same as the ambient light. Then,an effective color correction approach is designed to remove the color distortion; meanwhile,estimate the light source by the gray level algorithm. Next,the transmission fusion estimation method is established based on the transmission characteristics of water and the contrast constraint prior and is refined with multi-direction gradient weighted regularization. Finally,the clear image is obtained by solve the new underwater image model. The experimental results shows the image restored by the proposed underwater image restoration method which based on transmission fusion and optimization not only has the relatively natural color,but also maintains better detail and contrast information.
引文
[1]Li C,Guo J,Cong R,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.
[2]Huang B,Liu T,Hu H,et al.Underwater image recovery considering polarization effects of objects[J].Optics Express,2016,24(9):9826-9838.
[3]Peng Y,Cosman P.Underwater image restoration based on image blurriness and light absorption[J].IEEE Transactions on Image Processing,2017,26(4):1579-1594.
[4]Henke B,Vahl M,Zhou Z.Removing color cast of underwater images through non-constant color constancy hypothesis[C]//Proceedings of International Symposium on Image and Signal Processing and Analysis.Trieste,Italy,2014:20-24.
[5]Fu X,Zhuang P,Huang Y,et al.A retinex-based enhancing approach for single underwater image[C]//IEEEInternational Conference on Image Processing(ICIP).Paris,France,2014:4572-4576.
[6]Ancuti C,Ancuti C O,Haber T,et al.Enhancing underwater images and videos by fusion[C]//IEEEComputer Vision and Pattern Recognition(CVPR).Providence,USA,2012:81-88.
[7]He K,Sun J,Tang X.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI),2011,33(12):2341-2353.
[8]Drews P Jr,Nascimento E D,Moraes F,et al.Transmission estimation in underwater single images[C]//IEEE International Conference on Computer Vision Work-shops.Sydney,Australia,2013:825-830.
[9]Galdran A,Pardo D,Picon A,et al.Automatic redchannel underwater image restoration[J].Journal of Visual Communication and Image Representation,2015,26:132-145.
[10]De Oliveria G,Duarte J F,Moraes A,et al.Single image restoration for participating media based on prior fusion[J].IEEE Computer Graphics and Applications,2019,39:71-83.
[11]杨爱萍,田鑫,杨炳旺,等.基于多特征融合的单幅水下图像清晰化[J].天津大学学报:自然科学与工程技术版,2018,51(10):1031-1041.Yang Aiping,Tian Xin,Yang Bingwang,et al.Single underwater image sharpening based on multi-feature fusion[J].Journal of Tianjin University:Science and Technology,2018,51(10):1031-1041(in Chinese).
[12]Lu H,Li Y,Xu X,et al.Underwater image enhancement method using weighted guided trigonometric filtering and artificial light correction[J].Journal of Visual Communication and Image Representation,2016,38:504-516.
[13]Meng G,Wang Y,Duan J,et al.Efficient image dehazing with boundary constraint and contextual regularization[C]//IEEE International Conference on Computer Vision(ICCV).Sydney,Australia,2013:617-624.
[14]Fang F,Li F,Zeng T.Single image dehazing and denoising:A fast variational approach[J].Siam Journal on Imaging Sciences,2014,7(2):969-996.
[15]Li F,Wu J,Wang Y,et al.A color cast detection algorithm of robust performance[C]//IEEE 5th International Conference on Advanced Computational Intelligence(ICACI).Nanjing,China,2012:662-664.
[16]Yang J,Zhang Y.Alternating direction algorithms for l1-problems in compressive sensing[J].Siam Journal on Scientific Computing,2009,33(1):250-278.
[2]Huang B,Liu T,Hu H,et al.Underwater image recovery considering polarization effects of objects[J].Optics Express,2016,24(9):9826-9838.
[3]Peng Y,Cosman P.Underwater image restoration based on image blurriness and light absorption[J].IEEE Transactions on Image Processing,2017,26(4):1579-1594.
[4]Henke B,Vahl M,Zhou Z.Removing color cast of underwater images through non-constant color constancy hypothesis[C]//Proceedings of International Symposium on Image and Signal Processing and Analysis.Trieste,Italy,2014:20-24.
[5]Fu X,Zhuang P,Huang Y,et al.A retinex-based enhancing approach for single underwater image[C]//IEEEInternational Conference on Image Processing(ICIP).Paris,France,2014:4572-4576.
[6]Ancuti C,Ancuti C O,Haber T,et al.Enhancing underwater images and videos by fusion[C]//IEEEComputer Vision and Pattern Recognition(CVPR).Providence,USA,2012:81-88.
[7]He K,Sun J,Tang X.Single image haze removal using dark channel prior[J].IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI),2011,33(12):2341-2353.
[8]Drews P Jr,Nascimento E D,Moraes F,et al.Transmission estimation in underwater single images[C]//IEEE International Conference on Computer Vision Work-shops.Sydney,Australia,2013:825-830.
[9]Galdran A,Pardo D,Picon A,et al.Automatic redchannel underwater image restoration[J].Journal of Visual Communication and Image Representation,2015,26:132-145.
[10]De Oliveria G,Duarte J F,Moraes A,et al.Single image restoration for participating media based on prior fusion[J].IEEE Computer Graphics and Applications,2019,39:71-83.
[11]杨爱萍,田鑫,杨炳旺,等.基于多特征融合的单幅水下图像清晰化[J].天津大学学报:自然科学与工程技术版,2018,51(10):1031-1041.Yang Aiping,Tian Xin,Yang Bingwang,et al.Single underwater image sharpening based on multi-feature fusion[J].Journal of Tianjin University:Science and Technology,2018,51(10):1031-1041(in Chinese).
[12]Lu H,Li Y,Xu X,et al.Underwater image enhancement method using weighted guided trigonometric filtering and artificial light correction[J].Journal of Visual Communication and Image Representation,2016,38:504-516.
[13]Meng G,Wang Y,Duan J,et al.Efficient image dehazing with boundary constraint and contextual regularization[C]//IEEE International Conference on Computer Vision(ICCV).Sydney,Australia,2013:617-624.
[14]Fang F,Li F,Zeng T.Single image dehazing and denoising:A fast variational approach[J].Siam Journal on Imaging Sciences,2014,7(2):969-996.
[15]Li F,Wu J,Wang Y,et al.A color cast detection algorithm of robust performance[C]//IEEE 5th International Conference on Advanced Computational Intelligence(ICACI).Nanjing,China,2012:662-664.
[16]Yang J,Zhang Y.Alternating direction algorithms for l1-problems in compressive sensing[J].Siam Journal on Scientific Computing,2009,33(1):250-278.