Abstract：The attenuation of underwater image has a direct relation with the imaging distance.To restore the underwater image without accurate depth map,a segmented mapping approach based on scattering model is put forward.The scattering model of underwater light is studied and the attenuation law is explored on base of this model;the scattering model is built by linear fitting using pool test data;the segmented mapping approach based on scattering model is presented.To avoid finding out accurate depth map,discrete multiple-scale depth values are derived from mapping restrictions and a few prior knowledge.As a result,the function of segmented mapping from the degraded image to the actual image is built.Restoration results show that the approach is capable of enhancing the overall contrast,highlighting the detail,and improving the image quality effectively.
[1] Narasimhan S G.Models and algorithms for vision through the atmosphere[D].USA:Columbia University,2004.
[2] Narasimhan S G,Nayar S K.Interactive(De)weathering of an image using physical models[C]//IEEE Workshop on Color and Photometric Methods in Computer Vision.2003.
[3] 董慧颖,方帅,王欣威,等.基于物理模型的恶化天气下的图像复原方法及应用[J].东北大学学报:自然科学版,2005,26(3):217-219.Dong Huiying,Fang Shuai,Wang Xinwei,et al.A method based on physical model to restore degraded weather images and its application[J].Journal of Northeastern University:Natural Science,2005,26(3):217-219.
[4] 张瑞平.雾天条件下图像的恢复研究[D].天津:天津大学,2007.Zhang Ruiping.Research on image restoration in fog weather[D].Tianjin:Tianjin University,2007.
[5] 曹念文,张玉钧,王锋平,等.水下目标自然光偏振成像的研究[J].量子电子学报,1999,16(2):110-115.Cao Nianwen,Zhang Yujun,Wang Fengping,et al.White light polarization imaging for underwater objects[J].Chinese Journal of Quantum Electronics,1999,16(2):110-115.
[6] Schechner Y Y,Karpel N.Clear underwater vision[C]//IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Los Alamitos,CA,USA:IEEE Computer Society,2004:536-543.
[7] Queiroz-Neto J P,Carceroni R,Barros W,et al.Underwater stereo[C]//17th Brazilian Symposium on Computer Graphics and Image Processing.Los Alamitos,CA,USA:IEEE Computer Society,2004:170-177.
