图像去雾算法研究进展

doi:10.13229/j.cnki.jdxbgxb20200382

摘要/Abstract

摘要：

图像去雾技术根据算法的处理方式分为两类：一类是图像增强去雾算法，涉及直方图均衡化、小波变换和基于色彩恒常理论的Retinex算法等；另一类是图像复原去雾算法，主要包括传统基于光偏振特性的多幅图像复原和基于先验假设的单幅图像复原等方法，以及基于深度学习的图像复原算法。为了今后更好地研究图像去雾技术，本文系统地回顾了图像去雾算法的发展情况，对图像去雾算法研究中存在的问题进行了分析，并尝试探讨了其发展趋势。

关键词: 信息处理技术, 雾霾图像, 图像去雾, 图像增强, 图像复原

Abstract:

According to the algorithm processing method, image dehazing technology can be divided into two categories. One is image enhancement algorithm, which involves histogram equalization, wavelet transform and Retinex algorithm based on color constancy theory. The other is the image restoration and defogging algorithm, which mainly includes the traditional multi-image restoration based on the characteristics of optical polarization and the single image restoration based on a priori assumption, and the emerging image restoration algorithm based on deep learning. In order to better study the image dehazing algorithm in the future, the main development process of image dehazing techniques was reviewed in this paper, the existing problems for the study on image dehazing algorithm was analyzed, and the development trend was tried to explore.

Key words: information processing technology, fog-degraded image, image dehazing, image enhancement, image restoration

中图分类号:

TP391.4

蒋华伟,杨震,张鑫,董前林. 图像去雾算法研究进展[J]. 吉林大学学报(工学版), 2021, 51(4): 1169-1181.

Hua-wei JIANG,Zhen YANG,Xin ZHANG,Qian-lin DONG. Research progress of image dehazing algorithms[J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(4): 1169-1181.

图/表 7

图1

图2

图3

图4

图5

表1

典型图像去雾算法客观评价"

图像去雾算法	运算时间/s	$e$	$r ˉ$	$σ$
全局直方图均衡化	0.002 4	0.153 43	2.739	0.211 32
局部直方图均衡化	0.024 2	0.192 85	3.654	0.200 98
单尺度Retinex	0.044 3	0.227 37	5.256	0.219 67
多尺度Retinex	0.103 9	0.345 97	3.661	0.194 03
暗通道先验算法	0.135 8	0.411 58	2.825	0.127 57
中值滤波法	0.072 4	0.398 81	3.088	0.199 25
DehazeNet	0.891 3	0.415 39	2.461	0.115 39
DANet	0.338 1	0.399 26	2.523	0.105 93

表1

表2

典型图像去雾算法比较"

图像去雾算法	优点	缺点	适用情况	参考来源
全局直方图均衡化	算法简单、运算时间短、图像灰度离散程度高	合并较多灰度级，细节信息丢失，不适合景深多变的图像	单一景深图像	Jain^［57］，1989
局部直方图均衡化	整体对比度较强，能够保留图像细节信息	运算量大，存在局部块状效应	静态浓雾图像	Kim^［58］，1997；江巨浪等^［10］，2006
小波变换	可去除光照不均产生的黑斑，能较好地保持图像的原貌	占用较大的运算空间	薄雾、水下图像	Mallat^［59］，1989；吴颖谦^［14］，2003
局部单尺度Retinex（SSR）	弥补边缘曲线增强的局限性	产生光晕现象，整体亮度不足	薄雾图像	Jobson等^［19］，1997
局部多尺度Retinex（MSR）	相比SSR，图像处理后亮度得到提升	色彩饱和度不够，有轻度失真	环境亮度均匀的图像	汪荣贵等^［24］，2010
带色彩恢复的多尺度Retinex（MSRCR）	性能比较稳定，细节信息完善	算法复杂度增加，运算时间长	色彩失真的静态图像	Rahman等^［60］，1997
基于不同偏振度的多幅图像复原	能够消除偏振光的影响	依赖环境光的偏振状态和偏振度	薄雾图像，光线较好的图像	Schechner等^［30］，2001；Namer等^［31］，2005
暗通道先验	能够较大限度地保留图像细节信息	对大气光亮度估计较差，易造成细节缺失	景深连续且无遮挡的图像	He等^［35］，2011
中值滤波法	时间复杂度低，去雾性能高	物体边缘易产生块状效应	主体边缘简洁明确的图像	胡研等^［61］，2018
端到端去雾DehazeNet	利用神经网络高效估计退化模型参数	经特征提取等步骤，运算速度有限制	无杂乱纹理的自然场景图像	Cai等^［46］，2016
DANet	基于合成域和真实域，高效恢复雾图像	算法复杂度相对于其他神经网络算法较高	非极端环境的图像	Shao等^［55］，2020

表2

参考文献 61

1	Celik Turgay. Spatial entropy-based global and local image contrast enhancement[J].IEEE Transactions on Image Processing, 2014, 23(12): 5298-5308.
2	Oakley J P, Bu H. Correction of simple contrast loss in color images[J].IEEE Transactions on Image Processing, 2007, 16(2):511-522.
3	胡琼,汪荣贵,胡韦伟,等.基于直方图分割的彩色图像增强算法[J].中国图象图形学报, 2009, 14(9): 1776-1781.
	Hu Qiong, Wang Rong-gui, Hu Wei-wei, et al. Color image enhancement based on histogram segmentation[J]. Journal of Image and Graphics, 2009, 14(9):1776-1781.
4	张煜东,王水花,周振宇,等.基于HVS与PCNN的彩色图像增强[J].中国科学:信息科学, 2010, 40(7):909-924.
	Zhang Yu-dong, Wang Shui-hua, Zhou Zhen-yu, et al. Color image enhancement based on HVS and PCNN[J]. Science China(Information Sciences), 2010, 40(7): 909-924.
5	吴成茂.直方图均衡化的数学模型研究[J].电子学报, 2013, 41(3):598-602.
	Wu Cheng-mao. Studies on mathematical model of histogram equalization[J]. ACTA Electronica Sinica, 2013, 41(3):598-602.
6	Lee J S. Digital image enhancement and noise filtering by use of local statistics[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 1980, 2(2):165-168.
7	Kim T K, Paik J K, Kang B S. Contrast enhancement system using spatially adaptive histogram equalization with temporal filtering[J]. IEEE Transactions on Consumer Electronics, 1998, 44(1):82-87.
8	Kim J Y, Kim L S, Hwang S H. An advanced contrast enhancement using partially overlapped sub-block histogram equalization[J].IEEE Transactions on Circuits and Systems for Video Technology, 2001, 11(4):475-484.
9	曹聚亮,吕海宝,李冠章.基于自适应局部灰度修正的直方图均衡算法[J].红外与激光工程, 2004, 33(5): 513-515, 523.
	Cao Ju-liang, Lv Hai-bao, Li Guan-zhang. Histogram equalization algorithm base on adaptive gray level modification [J]. Infrared and Laser Engineering, 2004, 33(5): 513-515, 523.
10	江巨浪,张佑生,薛峰,等.保持图像亮度的局部直方图均衡算法[J].电子学报, 2006, 34(5): 861-866.
	Jiang Ju-lang, Zhang You-sheng, Xue Feng, et al. Local histogram equalization with brightness preservation[J].Acta Electronica Sinica, 2006,34(5):861-866.
11	陈文飞,廖斌,许雪峰,等.基于Piecewise直方图均衡化的图像增强方法[J].通信学报, 2011, 32(9): 153-160.
	Chen Wen-fei, Liao Bin, Xu Xue-feng, et al. Piecewise histogram equalization based image enhancement[J]. Journal on Communications, 2011, 32(9):153-160.
12	倪林.小波变换与图像处理[M].合肥:中国科学技术大学出版社, 2010.
13	陶观群,李大鹏,陆光华.基于小波变换不同融合规则的图像融合研究[J].红外与激光工程, 2003, 32(2): 173-176, 202.
	Tao Guan-qun, Li Da-peng, Lu Guang-hua. Study on image fusion based on different fusion rules of wavelet transform[J]. Infrared and Laser Engineering, 2003, 32(2):173-176, 202.
14	吴颖谦,施鹏飞.基于小波变换的低对比度图像增强[J].红外与激光工程, 2003, 32(1):4-7.
	Wu Ying-qian, Shi Peng-fei. Approach on image contrast enhancement based on wavelet transform[J]. Infrared and Laser Engineering, 2003, 32(1):4-7.
15	晁锐,张科,李言俊. 一种基于小波变换的图像融合算法[J].电子学报, 2004, 32(5):750-753.
	Chao Rui, Zhang Ke, Li Yan-jun. An image fusion algorithm using wavelet transform[J]. Acta Electronica Sinica, 2004, 32(5):750-753.
16	王智文,李绍滋. 基于多元统计模型的分形小波自适应图像去噪[J].计算机学报, 2014, 37(6):1380-1389.
	Wang Zhi-wen, Li Shao-zi. Adaptive fractal-wavelet image denoising base on multivariate statistical model[J]. Chinese Journal of Computers, 2014, 37(6):1380-1389.
17	李庆忠,刘清.基于小波变换的低照度图像自适应增强算法[J].中国激光, 2015,42(2):272-278.
	Li Qing-zhong, Liu Qing. Adaptive enhancement algorithm for low illumination images based on wavelet transform[J]. Chinese Journal of Lasers, 2015,42(2):272-278.
18	Land E H, Mccann J J. Lightness and retinex theory[J]. Journal of the Optical Society of America, 1971, 61(1):1-11.
19	Jobson D J, Rahman Z, Woodell G A. Properties and performance of a center/surround Retinex[J]. IEEE Transactions on Image Processig, 1997, 6(3):451-462.
20	占必超,吴一全,纪守新.基于平稳小波变换和Retinex的红外图像增强方法[J].光学学报, 2010, 30(10): 2788-2793.
	Zhan Bi-chao, Wu Yi-quan, Ji Shou-xin. Infrared image enhancement method based on stationary wavelet transformation and retinex[J]. ACTA Optica Sinica, 2010, 30(10):2788-2793.
21	马忠丽,文杰. 融合边缘信息的单尺度Retinex海雾去除算法[J].计算机辅助设计与图形学学报, 2015, 27(2):217-225.
	Ma Zhong-li, Wen Jie. Single-scale retinex sea fog removal algorithm fused the edge information[J]. Journal of Computer-Aided Design and Computer Graphics, 2015, 27(2): 217-225.
22	Kimmel R, Ellad M, Shaked D, et al. A variational framework for retinex[J]. International Journal of Computer Vision, 2003, 52(1):7-23.
23	Li M. A fast algorithm for color image enhancement with total variation regularization[J]. Science China(Information Sciences), 2010, 53(9):1913-1916.
24	汪荣贵,朱静,杨万挺,等.基于照度分割的局部多尺度Retinex算法[J].电子学报, 2010, 38(5):1181-1186.
	Wang Rong-gui, Zhu Jing, Yang Wan-ting, et al. An improved local multi-scale Retinex algorithm based on illumination image segmentation[J]. Acta Electronica Sinica, 2010, 38(5):1181-1186.
25	张尚伟,曾平,罗雪梅,等.具有细节补偿和色彩恢复的多尺度Retinex色调映射算法[J].西安交通大学学报,2012, 46(4): 32-37.
	Zhang Shang-wei, Zeng Ping, Luo Xue-mei, et al. Multi-scale Retinex with color restoration and detail compensation[J]. Journal of Xi'an Jiaotong University, 2012, 46(4):32-37.
26	智宁,毛善君,李梅.沙尘降质图像清晰化算法[J].中国图象图形学报, 2016, 21(12):1585-1592.
	Zhi Ning, Mao Shan-jun, Li Mei. Visibility restoration algorithm of dust-degraded images[J]. Journal of Image and Graphics, 2016, 21(12): 1585-1592.
27	刘海波,杨杰,吴正平,等.基于暗通道先验和Retinex理论的快速单幅图像去雾方法[J].自动化学报, 2015, 41(7):1264-1273.
	Liu Hai-bo, Yang Jie, Wu Zheng-ping, et al. A fast single image dehazing method based on dark channel prior and retinex theory[J]. Acta Automatica Sinica, 2015, 41(7):1264-1273.
28	Narasimhan S G, Nayar S K. Vision and the atmosphere[J]. International Journal of Computer Vision, 2002, 48(3):233-254.
29	Shurcliff W A, Ballard S S. Polarized Light[M]. Princeton: Van Nostrand, 1964.
30	Schechner Y Y, Narasimhan S G, Nayar S K. Instant dehazing of images using polarization[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Kauai, USA, 2001:325-332.
31	Namer E, Schechner Y Y. Advanced visibility improvement based on polarization filtered images[C]∥Proceedings of the Polarization Science and Remote Sensing,San Diego, USA,2005:36-45.
32	Schechner Y Y, Averbuch Y. Regularized image recovery in scattering media[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(9):1655-1660.
33	周浦城,薛模根,张洪坤,等. 利用偏振滤波的自动图像去雾[J].中国图象图形学报, 2011, 16(7):1178-1183.
	Zhou Pu-cheng, Xue Mo-gen, Zhang Hong-kun, et al. Automatic image dehaze using polarization filtering[J]. Journal of Image and Graphics, 2011, 16(7):1178-1183.
34	Oakley J P, Satherley B L. Improving image quality in poor visibility conditions using a physical model for contrast degradation[J]. IEEE Transactions on Image Processing, 1998, 7(2): 167-179.
35	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.
36	陈俊君.基于暗原色先验的图像去雾算法研究[D].沈阳: 东北大学机械工程与自动化学院, 2013.
	Chen Jun-jun. Research on image defogging based on dark channel prior[D]. Shenyang: School of Mechanical Engineering and Automation,Northeastern University, 2013.
37	张选德,冯象初,王卫卫,等.基于双向非局部模型的图像插值[J].中国科学:技术科学, 2013, 43(5):564-573.
	Zhang Xuan-de, Feng Xiang-chu, Wang Wei-wei, et al. Two-direction nonlocal model for image interpolation[J]. Science China(Technology Sciences), 2013, 43(5):564-573.
38	张小刚,唐美玲,陈华,等.一种结合双区域滤波和图像融合的单幅图像去雾算法[J].自动化学报, 2014, 40(8):1733-1739.
	Zhang Xiao-gang, Tang Mei-ling, Chen Hua, et al. A dehazing method in single image based on double-area filter and image fusion[J]. Acta Automatica Sinica, 2014, 40(8):1733-1739.
39	杨燕,陈高科.基于光补偿和逐像素透射率的图像复原算法[J].通信学报, 2017,38(5):48-56.
	Yang Yan, Chen Gao-ke. Single image visibility restoration using optical compensation and pixel-by-pixel transmittance[J]. Journal on Communications, 2017, 38(5):48-56.
40	曲晨, 毕笃彦. 基于懒惰随机游走的雾天含噪图像清晰化[J].光学学报, 2018, 38(4):95-104.
	Qu Chen, Bi Du-yan. Clearness for foggy and noisy image based on lazy random walk[J]. ACTA Optica Sinica,2018, 38(4): 95-104.
41	杨燕, 王志伟. 基于均值不等关系优化的自适应图像去雾算法[J]. 电子与信息学报, 2020, 42(3): 755-763.
	Yang Yan, Wang Zhi-wei. Adaptive image dehazing algorithm based on mean unequal relation optimization[J]. Journal of Electronics and Information Technology, 2020, 42(3): 755-763.
42	Iwamoto R, Yutaro H, Hashimoto M, et al. Real time haze removal using normalised pixel wise dark channel prior and robust atmospheric light estimation[J]. Applied Sciences,2020,10(3): 1165-1176.
43	Fattal R. Single image dehazing[J]. ACM Transaction on Graphics, 2008, 27(3):1-9.
44	Tan R T. Visibility in bad weather from a single image[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, USA, 2008: 2347-2354.
45	梅英杰,宁媛,陈进军.融合暗通道先验和MSRCR的分块调节图像增强算法[J].光子学报,2019,48(7): 124-135.
	Mei Ying-jie, Ning Yuan, Chen Jin-jun. Block-adjusted image enhancement algorithm combining dark channel prior with MSRCR[J]. Acta Photonica Sinica, 2019,48(7): 124-135.
46	Cai B L, Xu X M, Jia K, et al. DehazeNet: an end-to-End system for single image haze removal[J]. IEEE Transactions on Image Processing, 2016, 25(11):5187-5198.
47	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): 149-158.
	Chen Yong, Guo Hong-guang, Ai Ya-peng. Single image dehazing method based on multi-scale convolution neural network[J]. ACTA Optica Sinica, 2019,39(10):149-158.
48	刘杰平,杨业长,陈敏园,等.结合卷积神经网络与动态环境光的图像去雾算法[J]. 光学学报, 2019,39(11): 120-131.
	Liu Jie-ping, Yang Ye-chang, Chen Min-yuan, et al. Image dehazing algorithm based on convolutional neural network and dynamic ambient light[J]. Acta Optica Sinica, 2019, 39(11):120-131.
49	Li B Y, Peng X L, Wang Z Y. An all-in-one network for dehazing and beyond[C]∥IEEE International Conference on Computer Vision, Venice, Italy,2017:1-12.
50	Zhang H, Patel V M. Densely connected pyramid dehazing network[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, USA, 2018: 3194-3203.
51	Chen D D, He M M, Fan Q N. Gated context aggregation network for image dehazing and deraining[C]∥IEEE Conference on Applications of Computer Vision, Hawaii,USA,2019:1375-1383.
52	Zhang X Q,Wang T, Wang J X,et al. Pyramid channel based feature attention network for image dehazing[J]. Computer Vision and Image Understanding, 2020, 197-198:101003.
53	Li Z H,Gui B,Zhen T,et al. Grain depot image dehazing via quadtree decomposition and convolutional neural networks[J]. Alexandria Engineering Journal, 2020, 59(3):1463-1472.
54	Qin X,Wang Z L,Bai Y C, et al. FFA-Net: feature fusion attention network for single image dehazing[J]. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(7):11908-11915.
55	Shao Y J, Li L, Ren W Q, et al. Domain adaptation for image dehazing[C]∥IEEE Conference on Computer Vision and Pattern Recognition, Seattle, USA,2020: 2808-2817.
56	Hautiere N, Tarel J P, AUBERT D, et al. Blind contrast enhancement assessment by gradient ratioing at visible edges[J]. Image Analysis & Stereology Journal, 2008,27(2): 87-95.
57	Jain A K. Fundamentals of digital image processing[M]. Englewood Cliffs: Prentice Hall, 1989.
58	Kim Y T. Contrast enhancement using brightness preserving bi-histogram equalization[J]. IEEE Transactions on Consumer Electronics, 1997, 43(1):1-8.
59	Mallat S G. A theory for multi resolution signal decomposition: the wavelet representation[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1989, 11(7): 674-693.
60	Rahman Z, Jobson D J, Woodell G A. A multiscale retinex for bridging the gap between color images and the human observation of scenes[J]. IEEE Transactions on Image Processing, 1997, 6(7): 965-976.
61	胡妍,王柯俨,许宁,等.利用分割中值滤波和透射率补偿的图像去雾[J]. 西安电子科技大学学报:自然科学版,2018, 45(4):99-105.
	Hu Yan, Wang Ke-yan, Xu Ning,et al. Image dehazing by the segmenting median filter and transmission compensation[J]. Journal of Xidian University (Natural Science), 2018, 45(4): 99-105.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed

[1]	金静,党建武,王阳萍,申东. 融合模糊统计纹理特征的多线索粒子滤波跟踪[J]. 吉林大学学报(工学版), 2021, 51(3): 1111-1120.
[2]	郭继昌,乔珊珊. 基于深度图的水下图像复原[J]. 吉林大学学报(工学版), 2021, 51(2): 677-684.
[3]	刘富,刘璐,侯涛,刘云. 基于优化MSR的夜间道路图像增强方法[J]. 吉林大学学报(工学版), 2021, 51(1): 323-330.
[4]	刘国华,周文斌. 基于卷积神经网络的脉搏波时频域特征混叠分类[J]. 吉林大学学报(工学版), 2020, 50(5): 1818-1825.
[5]	史再峰,李金卓,曹清洁,李慧龙,胡起星. 基于生成对抗网络的低剂量能谱层析成像去噪算法[J]. 吉林大学学报(工学版), 2020, 50(5): 1755-1764.
[6]	王柯俨,王迪,赵熹,陈静怡,李云松. 基于卷积神经网络的联合估计图像去雾算法[J]. 吉林大学学报(工学版), 2020, 50(5): 1771-1777.
[7]	谌华,郭伟,闫敬文,卓文浩,吴良斌. 基于深度学习的SAR图像道路识别新方法[J]. 吉林大学学报(工学版), 2020, 50(5): 1778-1787.
[8]	张薇,韩勇,金铭,乔晓林. 基于托普利兹矩阵集重构的相干信源波达方向估计[J]. 吉林大学学报(工学版), 2020, 50(2): 703-710.
[9]	程艳芬,姚丽娟,袁巧,陈先桥. 水下视频图像清晰化方法[J]. 吉林大学学报(工学版), 2020, 50(2): 668-677.
[10]	于晓辉,张志成,李新波,孙晓东. 基于状态空间模型的指数衰减正弦信号参数估计[J]. 吉林大学学报(工学版), 2019, 49(6): 2083-2088.
[11]	刘富, 权美静, 王柯, 刘云, 康冰, 韩志武, 侯涛. 仿蝎子振源定位机理的位置指纹室内定位方法[J]. 吉林大学学报(工学版), 2019, 49(6): 2076-2082.
[12]	马子骥,卢浩,董艳茹. 双通道单图像超分辨率卷积神经网络[J]. 吉林大学学报(工学版), 2019, 49(6): 2089-2097.
[13]	郭继昌,吴洁,郭春乐,朱明辉. 基于残差连接卷积神经网络的图像超分辨率重构[J]. 吉林大学学报(工学版), 2019, 49(5): 1726-1734.
[14]	曹运合,曾丽,王宇. 基于特征空间的子阵级自适应和差波束测角方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1735-1744.
[15]	卢洋,王世刚,赵文婷,赵岩. 基于离散Shearlet类别可分性测度的人脸表情识别方法[J]. 吉林大学学报(工学版), 2019, 49(5): 1715-1725.