﻿ 一种基于稀疏系数匹配学习的图像去雾算法
 控制与决策  2020, Vol. 35 Issue (11): 2797-2802 0

### 引用本文 [复制中英文]

[复制中文]
NAN Dong, WANG Zhi-tian, ZHENG Shao-hua, HE Lin-yuan. An image dehazing method based on learning framework with sparse coefficient matching[J]. Control and Decision, 2020, 35(11): 2797-2802. DOI: 10.13195/j.kzyjc.2018.1764.
[复制英文]

### 文章历史

1. 陆军装甲兵学院 蚌埠校区，安徽 蚌埠 233050;
2. 空军工程大学 航空工程学院，西安 710038

An image dehazing method based on learning framework with sparse coefficient matching
NAN Dong 1, WANG Zhi-tian 1, ZHENG Shao-hua 1, HE Lin-yuan 2
1. Bengbu Campus, Academy of Army Armored Forces, Bengbu~233050, China;
2. Institute of Aeronautics, Air Force Engineering University, Xi'an 710038, China
Abstract: Due to the low accuracy of the existing image dehazing methods with prior, an image dehazing method based on a learning framework with sparse coefficient matching is proposed. Firstly, the solution of the hazy degradation model is transformed to sparse coefficient matching with the database from the view of image restoration. Then, to improve the visual effect of the result, a feedback iteration is quantified by the enhancement of the contrast in highlighted areas from the view of image enhancement. Experiments demonstrate that the proposed method can remove effectively haze as well as provide a good local detail, and it has good generality.
Keywords: image dehazing    hazy degradation model    sparse representation    learning framework
0 引言

1 本文算法基础 1.1 雾天退化模型

McCartney模型揭示了复杂气象条件下雾天图像的产生机理, 是各类算法求解的基础, 在文献[8, 11]不断优化的基础上得到如下所示的雾天退化模型:

 (1)

1.2 图像稀疏表示

 (2)
 图 1 图像稀疏表示

1.3 本文算法可行性分析

2 本文算法实现

 图 2 算法框架
2.1 模拟雾天图像数据库构建

 (3)

2.2 基于最大相关稀疏系数匹配的大气传递图生成

 (4)

 (5)

2.3 基于反馈迭代的去雾图像生成

 图 3 图像去雾结果

 图 4 去雾曲线(A = 255)

 图 5 反馈迭代图像去雾结果

3 实验结果及分析

3.1 数据库性能验证

 图 6 大气传递的数据库实验结果

3.2 雾天图像实验

 图 7 “人群”去雾结果
 图 8 “建筑”去雾结果
 图 9 “城市”去雾结果
 图 10 “山脉”去雾结果

4 结论

 [1] 吴迪, 朱青松. 图像去雾的最新研究进展[J]. 自动化学报, 2015, 41(2): 221-239. (Wu D, Zhu Q S. The last research progress of image dehazing[J]. Acta Automatica Sinica, 2015, 41(2): 221-239.) [2] Li Z G, Zheng J H. Edge-preserving decomposition based single image haze removal[J]. IEEE Transactions on Image Processing, 2015, 24(12): 5432-5441. [3] Kim K, Kim S, Kim K S. Effective image enhancement techniques for fog-affected indoor and outdoor images[J]. IET Image Processing, 2018, 12(4): 465-471. [4] Li Z G, Zheng J H. Single image de-hazing using globally guided image filtering[J]. IEEE Transactions on Image Processing, 2018, 27(1): 1284-1294. [5] McCartney E J. Optics of atmosphere: Scattering by molecules and particles[M]. New York: John Wiley and Sons, 1976: 23-32. [6] Alajarmeh A, Salam R A, Abdulrahim K, et al. Real-time framework for image dehazing based on linear transmission and constant-time airlight estimation[J]. Information Sciences, 2018, 436(4): 108-130. [7] Tan R T. Visibility in bad weather from a single image[C]. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York: IEEE Computer Society, 2008: 2347-2354. [8] He K M, Sun J, Tang X O. Single image haze removal using dark channel prior[C]. Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami: IEEE Computer Society, 2009: 1956-1963. [9] Nishino K, Kratz L, Lombardi S. Bayesian defogging[J]. International Journal of Compute Vision, 2012, 98(3): 263-278. [10] Trung M B, Kim W. Single image dehazing using color ellipsoid prior[J]. IEEE Signal Processing Society, 2018, 27(2): 999-1009. [11] Narasimhan S G, Nayar S K. Contrast restoration of weather degraded images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(6): 713-724. [12] 焦莉娟, 王文剑, 赵青杉, 等. 近邻局部OMP稀疏表示图像去噪[J]. 中国图象图形学报, 2018, 22(11): 1486-1492. (Jiao L J, Wang W J, Zhao Q S, et al. Image denoising based on sparse representation of neighbor local OMP[J]. Journal of Image and Graphics, 2018, 22(11): 1486-1492.) [13] Yang X, Sun Q, Wang T. No-reference image quality assessment based on sparse representation[J]. Neural Computing and Applications, 2018, 31(9): 1-16. [14] Ling Z, Fan G L, Gong J W, et al. Learning deep transmission network for efficient image dehazing[J]. Multimedia Tools and Applications, 2018, 18(11): 1-24. [15] Geng L, Sun Q S, Peng F, et al. Multi-scale fractional-order sparse representation for image denoising[J]. IEEE / CAA Journal of Automatica Sinica, 2018, 5(2): 1-10. [16] Tang K, Yang J C, Wang J. Investigating haze-relevant features in a learning framework for image dehazing[C]. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus: IEEE, 2014: 2995-3002. [17] Hou F, Jin C, Dong G. Weak fault feature extraction of rolling bearings based on globally optimized spare coding and approximate SVD[J]. Mechanical Systems and Signal Processing, 2018, 111(10): 234-250. [18] He K M, Sun J, Tang X. Guided image filtering[C]. Proceedings of the 2011 European Conference on Computer Vision. Berlin: Springer Heidelberg, 2011: 1-14. [19] Ren W, Liu S, Zhang H, et al. Single image dehazing via multi-scale convolutional neural networks[C]. Proceedings of the 2016 European Conference on Computer Vision. Amsterdam: Springer International Publishing, 2016: 154-169. [20] Berman D, Treibitz T, Avidan S. Non-local image dehazing[C]. Proceedings of the 2019 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas: IEEE, 2016: 1674-1682. [21] Mittal A, Soundararajan R, Bovik A C. Making a "completely blind" image quality analyzer[J]. IEEE Signal Processing Letters, 2013, 20(3): 209-212.