首页 > 论文 > 光学学报 > 39卷 > 10期(pp:1010001--1)

基于多尺度卷积神经网络的单幅图像去雾方法

Single Image Dehazing Method Based on Multi-Scale Convolution Neural Network

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

雾是一种常见的自然天气现象。雾天场景下的空气中悬浮着大量的微小水滴,这些水滴的散射和折射作用会使成像系统获得的图像偏灰白色,导致图像的色彩饱和度和对比度下降,从而丢失很多重要的细节信息,不利于图像特征的提取和辨识,也增加了对图像进行后续处理的难度。因此,研究如何对雾天场景下获得的退化图像进行有效处理,对大气退化图像的复原和景物细节信息的增强有着非常重要的现实意义和广阔的应用前景[1-2]。此外,随着对计算机视觉领域研究的不断深入,人们对成像设备采集的图像清晰度有了更高的要求,雾天图像清晰化已经成为计算机视觉的重要研究内容[3]。

Abstract

Since the traditional single image dehazing algorithm is susceptible to the prior knowledge constraint of hazy image and color distortion, this paper proposes a multi-scale convolutional neural network (CNN) single image dehazing method based on deep learning, which realizes image dehazing by learning the mapping relationship between hazy image and atmospheric transmission. According to the hazy image forming mechanism of atmospheric scattering model, an end-to-end multi-scale full CNN model is designed. The shallow layer features of hazy image are extracted by convolution layer operation, and then the deep features are extracted by multi-scale convolution kernel in parallel. Then the shallow layer features and deep features are fused by jump connection. Finally, the non-linear regression method is used to obtain the corresponding transmission features of the hazy image. According to the atmospheric scattering model, the haze-free image is restored. The model is trained by using hazy image data sets. The experimental results show that the proposed method can achieve good dehazing effect in the experiments of synthesizing hazy images and real natural hazy images. The proposed method is superior to other contrast algorithms in subjective and objective evaluations.

Newport宣传-MKS新实验室计划
补充资料

DOI:10.3788/AOS201939.1010001

所属栏目:图像处理

基金项目:国家自然科学基金、长江学者和创新团队发展计划、教育部人文社会科学研究基金;

收稿日期:2019-04-28

修改稿日期:2019-06-03

网络出版日期:2019-10-01

作者单位    点击查看

陈永:兰州交通大学电子与信息工程学院, 甘肃 兰州 730070
郭红光:兰州交通大学电子与信息工程学院, 甘肃 兰州 730070
艾亚鹏:兰州交通大学电子与信息工程学院, 甘肃 兰州 730070

联系人作者:陈永(edukeylab@126.com)

备注:国家自然科学基金、长江学者和创新团队发展计划、教育部人文社会科学研究基金;

【1】Wu D and Zhu Q S. The latest research progress of image dehazing. Acta Automatica Sinica. 41(2), 221-239(2015).
吴迪, 朱青松. 图像去雾的最新研究进展. 自动化学报. 41(2), 221-239(2015).

【2】Xu Y, Wen J, Fei L K et al. Review of video and image defogging algorithms and related studies on image restoration and enhancement. IEEE Access. 4, 165-188(2016).

【3】Ma R Q and Zhang S J. An improved color image defogging algorithm using dark channel model and enhancing saturation. Optik. 180, 997-1000(2019).

【4】Liu D M and Chang F L. Coarse-to-fine saliency detection based on non-subsampled contourlet transform enhancement. Acta Optica Sinica. 39(1), (2019).
刘冬梅, 常发亮. 基于非下采样轮廓小波变换增强的从粗到精的显著性检测. 光学学报. 39(1), (2019).

【5】Ancuti C O and Ancuti C. Single image dehazing by multi-scale fusion. IEEE Transactions on Image Processing. 22(8), 3271-3282(2013).

【6】He L Y, Zhao J Z, Zheng N N et al. Haze removal using the difference-structure-preservation prior. IEEE Transactions on Image Processing. 26(3), 1063-1075(2017).

【7】He K M, Sun J and Tang X O. Single image haze removal using dark channel prior. [C]//2009 IEEE Conference on Computer Vision and Pattern Recognition, June 20-25, 2009, Miami, FL, USA. New York: IEEE. 1956-1963(2009).

【8】He K M, Sun J and Tang X O. Guided image filtering. IEEE Transactions on Pattern Analysis and Machine Intelligence. 35(6), 1397-1409(2013).

【9】Jiang J L, Sun W, Wang Z D et al. Integrated enhancement algorithm for hazy image using transmittance as weighting factor. Journal of Electronics & Information Technology. 40(10), 2388-2394(2018).
江巨浪, 孙伟, 王振东 等. 基于透射率权值因子的雾天图像融合增强算法. 电子与信息学报. 40(10), 2388-2394(2018).

【10】Lu H B, Zhao Y F, Zhao Y J et al. Image defogging based on combination of image bright and dark channels. Acta Optica Sinica. 38(11), (2018).
卢辉斌, 赵燕芳, 赵永杰 等. 基于亮通道和暗通道结合的图像去雾. 光学学报. 38(11), (2018).

【11】Meng G F, Wang Y, Duan J Y et al. Efficient image dehazing with boundary constraint and contextual regularization. [C]//2013 IEEE International Conference on Computer Vision, December 1-8, 2013, Sydney, Australia. New York: IEEE. 617-624(2013).

【12】Tang K T, Yang J C and Wang J. Investigating haze-relevant features in a learning framework for image dehazing. [C]//2014 IEEE Conference on Computer Vision and Pattern Recognition, June 23-28, 2014, Columbus, OH, USA. New York: IEEE. 2995-3002(2014).

【13】Cai B L, Xu X M, Jia K et al. DehazeNet: an end-to-end system for single image haze removal. IEEE Transactions on Image Processing. 25(11), 5187-5198(2016).

【14】Ren W Q, Liu S, Zhang H et al. Single image dehazing via multi-scale convolutional neural networks. //Leibe B, Mata S, Sebe N, et al. Lecture notes in computer science. Cham: Springer. 9906, 154-169(2016).

【15】Cox L J. Optics of the atmosphere-scattering by molecules and particles. Optica Acta: International Journal of Optics. 24(7), (1977).

【16】Li B Y, Ren W Q, Fu D P et al. -08-27)[2019-03-10]. (2018).

引用该论文

Yong Chen,Hongguang Guo,Yapeng Ai. Single Image Dehazing Method Based on Multi-Scale Convolution Neural Network[J]. Acta Optica Sinica, 2019, 39(10): 1010001

陈永,郭红光,艾亚鹏. 基于多尺度卷积神经网络的单幅图像去雾方法[J]. 光学学报, 2019, 39(10): 1010001

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF