[1]潘健鸿,高 银.基于天空区域分割和多尺度融合的单幅雾天图像复原算法[J].南京理工大学学报(自然科学版),2019,43(05):592-599.[doi:10.14177/j.cnki.32-1397n.2019.43.05.008]
 Pan Jianhong,Gao Yin.Single image dehazing with sky region segmentationand multi-scale fusion[J].Journal of Nanjing University of Science and Technology,2019,43(05):592-599.[doi:10.14177/j.cnki.32-1397n.2019.43.05.008]
点击复制

基于天空区域分割和多尺度融合的单幅雾天图像复原算法()
分享到:

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

卷:
43卷
期数:
2019年05期
页码:
592-599
栏目:
出版日期:
2019-10-31

文章信息/Info

Title:
Single image dehazing with sky region segmentationand multi-scale fusion
文章编号:
1005-9830(2019)05-0592-08
作者:
潘健鸿1高 银2
1.福建省特种设备检验研究院,福建 福州 350008; 2.中国科学院 泉州装备制造研究所,福建 泉州 362200
Author(s):
Pan Jianhong1Gao Yin2
1.Fujian Special Equipment Inspection and Research Institute,Fuzhou 350008,China; 2.Quanzhou Institute of Equipment Manufacturing,Chinese Academy of Sciences,Quanzhou 362200,China
关键词:
图像去雾 天空区域分割 多尺度融合 L0梯度最小化 暗通道算法
Keywords:
image dehazing sky region segmentation multi-scale fusion L0 gradient minimization dark channel theory
分类号:
TP751; TP391
DOI:
10.14177/j.cnki.32-1397n.2019.43.05.008
摘要:
传统的暗原色理论相关算法在处理带有天空区域的雾天图像时易出现亮度损失和光晕现象。针对这一问题,该文提出一种基于空区域分割和多尺度融合的单幅雾天图像复原算法。该方法首先对图像进行缩放,根据雾天图像的特性分割天空区域,获取全局大气背景光; 根据散射模型获取初步的透射率图像,运用L0梯度最小化方法获取优化的透射率图像; 最后运用多尺度融合的方法对不同曝光度的图像进行融合,获取最终的去雾图像。主观观察和客观评价表明,在整体和细节方面,该算法比现有暗原色算法及其改进算法处理效果更好。
Abstract:
To deal with the problems of hue phenomenon and brightness distortion in the related dark channel theory algorithm,the single fog image restoration algorithm with the sky region segmentation and multi-scale fusion is proposed. Firstly,the global atmospheric light is effectively obtained in the sky regions with characteristics of scaled hazy images. To properly optimize the transmission,an initial dark channel image is contructed,and the L0 gradient minimization method is used to obtain the optimized transmission image. Finally,the fusion theory is applied to further modify the transmittance image,and the dehazing image is obtained. The subjective observation and objective evaluation show that,the algorithm is better than the classic dark channel algorithm and the improved algorithm in the overall and details.

参考文献/References:

[1] Yu J,Liao Q. Fast single image fog removal using edge-preserving smoothing[C]//IEEE International Conference on Acoustics,Speech and Signal Processing. Prague,Czech Republic:IEEE,2011:1245-1248.
[2]陈功,王唐,周荷琴. 基于物理模型的雾天图像复原新方法[J]. 中国图像图形学报,2008,13(5):888-893.
Chen Gong,Wang Tang,Zhou Heqin. A novel physics based method for restoration of foggy day images[J]. Journal of Image and Graphics,2008,13(5):888-893.
[3]Land E H,Mccann J J. Lightness and retinex theory[J]. Journal of the Optical Society of America,1971,61(1):1-11.
[4]Brainard D H,Wandell B A. Analysis of the retinex theory of color vision[J]. Opt Soc Amer,1986,3(10):1651-1661.
[5]He K,Sun J,Tang X. Single image haze removal using dark channel prior[C]//2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami,US:IEEE Conference on CVPR,2009:1956-1963.
[6]Nayar S K,Narasimhan S G. Vision in bad weather[C]//IEEE International Conference on Computer Vision. Kerkyra,Greece:IEEE,2002:820-827.
[7]Fattal R. Single image dehazing[J]. ACM Trans Graph,2008,27(3):1-9.
[8]Meng G,Wang Y,Duan J,et al. Efficient image dehazing with boundary constraint and contextual regularization[C]//IEEE International Conference on Computer Vision. Sydney,Australia:IEEE,2014:617-624.
[9]高银,云利军,石俊生,等. 基于TV模型的暗原色理论雾天图像复原算法[J]. 中国激光,2015(8):273-278.
Gao Yin,Yun Lijun,Shi Junsheng,et al. Enhancement dark channel algorithm of fog image based on the TV Model[J]. Chinese Journal of Lasers,2015,39(5):544-549.
[10]高银,云利军,石俊生,等. 基于四阶PDE模型的暗原色理论雾天图像增强算法[J]. 南京理工大学学报,2015,39(5):544-549.Gao Yin,Yun Lijun,Shi Junsheng,et al. Enhancement dark channel theory algorithm of fog image based on fourth-order PDE model[J]. Journal of Nanjing University of Science and Technology,2015,39(5):544-549.
[11]Guo Fan,Tang Jin,Cai Zixing. Fusion strategy for single image dehazing[C]//2015 Global Conference on Communication Technologices(GCCT). Thuckalay,India:Communication Technologies(GCCT),2015:580-584.
[12]Gao Y,Su Y,Li Q,et al. Single fog image restoration with multi-focus image fusion[J]. Journal of Visual Communication and Image Representation,2018,55:586-595.
[13]He K,Sun J,Tang X. Guided image filtering[C]//11th European Conference on Computer Vision(ECCV). Crete,Greece:European Conference on Computer Vision,2010:1-14.
[14]Xu L,Lu C,Xu Y,et al. Image smoothing via L0 gradient minimization[J]. ACM Transactions on Graphics,2011,30(6):1-12.
[15]Sujoy Paul,Ioana S Sevcenco,Panajotis Agathoklis. Multi-exposure and multi-focus image fusion in gradient domain[J]. Journal of Circuits Systems & Computers,2016,25(10):1650123.
[16]Sulami M,Glatzer I,Fattal R,et al. Automatic recovery of the atmospheric light in hazy images[C]//2014 IEEE International Conference on Computational Photography. Santa Clara,US:IEEE,2014:1-11.
[17]Berman D,Treibitz T,Avidan S. Non-local image dehazing[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. NV,US:IEEE,2016:1674-1682.
[18]Hautière N,Tarel J P,Aubert D,et al. Blind contrast enhancement assessment by gradient ratioing at visible edges[J]. Image Analysis & Stereology,2011,27(2):87-95.
[19]Fang Y,Ma K,Wang Z,et al. No-reference quality assessment of contrast-distorted images based on natural scene statistics[J]. IEEE Signal Processing Letters,2015,22(7):838-842.

备注/Memo

备注/Memo:
收稿日期:2018-10-17 修回日期:2019-03-09
基金项目:质检总局科技计划项目(2017QK151)
作者简介:潘健鸿(1979-),男,硕士,主要研究方向:智能制造、数据分析等,E-mail:fjtjpanjh@foxmail.com; 通讯作者:高银(1985-),男,硕士,工程师,主要研究方向:图像复原、增强及融合等,E-mail:yngaoyin@163.com。
引文格式:潘健鸿,高银. 基于天空区域分割和多尺度融合的单幅雾天图像复原算法[J]. 南京理工大学学报,2019,43(5):592-599.
投稿网址:http://zrxuebao.njust.edu.cn
更新日期/Last Update: 2019-11-30