发布时间: 2016-09-25 摘要点击次数： 全文下载次数： DOI: 10.11834/jig.20160911 2016 | Volumn 21 | Number 9 图像理解和计算机视觉

 收稿日期: 2015-12-30; 修回日期: 2016-04-22 基金项目: 国家自然科学基金项目（61003188, 61379075）；浙江省自然科学基金项目（LY14F020004）；国家科技支撑计划项目（2014BAK14B01）；浙江省公益性技术应用研究计划项目（2015C33071）；浙江工商大学青年人才基金项目（QZ13-9）；浙江省智能交通工程技术研究中心开放课题（2015ERCITZJ-KF1） 第一作者简介: 赵锦威(1995-), 男, 现为浙江工商大学计算机科学与技术专业本科生, 主要研究方向为图像视频处理与增强复原技术。E-mail:clarkzjw@gmail.com 中图法分类号: TP391.4 文献标识码: A 文章编号: 1006-8961(2016)09-1221-08

# 关键词

Dark channel prior-based image dehazing with atmospheric light validation and halo elimination
Zhao Jinwei, Shen Yiyun, Liu Chunxiao, Ouyang Yi
School of Computer Science & Information Engineering, Zhejiang Gongshang University, Hangzhou 310018, China
Supported by: National Natural Science Foundation of China(61003188, 61379075); Natural Science Foundation of Zhejiang Province, China(LY14F020004)

# Abstract

Objective To address false candidate atmospheric light and halo effects in dark channel prior-based image dehazing methods, an improved image dehazing algorithm is proposed, with atmospheric light validation and halo elimination strategies, which can reveal high visibility in haze-free images. Method As dark channel prior-based methods do not verify the validity of atmospheric light and may fail to select a proper one, a support vector machine-based classifier is trained and utilized to reject the false candidate atmospheric light. The halo effect is introduced to haze free results with coarse transmission maps in dark channel prior-based approaches. Thus, a patch shift-based fine transmission estimation method is adopted, which can preserve edges in the input image and suppress halo effects in the haze-free image significantly. Several pixels may still remain with halo effects near sharp edges, which are detected and corrected by the proposed halo elimination strategy using the guided filter. Finally, the haze-free image is obtained by solving the haze image formation model. Result Experimental results demonstrate that the false candidate atmospheric light is rejected and the halo effect is diminished significantly when our algorithm is applied. The resulting haze-free images possess superior visibility, rich image details, and depth. Conclusion Our algorithm outperforms the state-of-the-art image dehazing methods and significantly improves visibility in haze-free images, which meets the requirements of applications such as video surveillance, traffic navigation, and object detection.

# Key words

image dehazing; dark channel prior; patch shift; atmospheric light; halo elimination

# 1 本文算法

 $\boldsymbol{I}\left( x \right) = \boldsymbol{J}\left( x \right)t\left( x \right) + \boldsymbol{A}\left( {1 - t\left( x \right)} \right)\;$ (1)

 $\boldsymbol{J}\left( x \right)= \frac{{\boldsymbol{I}\left( x \right) - \boldsymbol{A}}}{{\max \left( {t\left( x \right),{t_0}} \right)}} + \boldsymbol{A}$ (2)

t0为防止分母为0引入的阈值，一般取值为0.1，令$k = \frac{1}{{\max \left( {t\left( x \right),{t_0}} \right)}}$, 则

 $\boldsymbol{J}\left( x \right) = k\boldsymbol{I}\left( x \right) + \left( {1 - k} \right)\boldsymbol{A}\;$ (3)

# 1.1 基于SVM的误判大气光校验方法

 ${\boldsymbol{I}^{{\rm{dark}}}}\left( y \right) = \mathop {\min }\limits_{y \in \Omega \left( x \right)} \left( {\mathop {\min }\limits_c \frac{{{I^c}\left( y \right)}}{{{A^c}}}} \right)\;$ (4)

 $\begin{array}{*{20}{c}} {\mathop {\min }\limits_{w,b,x} \frac{1}{2}{\boldsymbol{w}^{\rm{T}}}\boldsymbol{w} + C\sum\limits_{i = 1}^l {{\xi _i}} }\\ {s.t.\;\;\;{y_i}\left( {{\boldsymbol{w}^{\rm{T}}}\varphi \left( {{X_i}} \right) + b} \right) \ge 1 - {\xi _i}}\\ {{\xi _i} \ge 0,i = 1, \cdots ,l} \end{array}\;$ (5)

# 1.2 基于块偏移的精确透射率计算策略

 $\boldsymbol{\widetilde t}\left( x \right) = 1 - \mathop {\min }\limits_{y \in \Omega \left( x \right)} \left( {\mathop {\min }\limits_c \frac{{{I^c}\left( y \right)}}{{{A^c}}}} \right)\;$ (6)

 $P\left( {\frac{{{I^c}\left( x \right)}}{{{A^c}}}} \right) = \mathop {\min }\limits_{y \in \Omega \left( x \right)} \left( {\mathop {\min }\limits_c \frac{{{I_0}\left( y \right)}}{{{A^c}}}} \right)$ (7)

 ${t_1}\left( x \right) = 1 - P\left( {\frac{{{I^c}\left( x \right)}}{{{A^c}}}} \right)$ (8)

# 参考文献

• [1] 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. DOI:10.1109/TPAMI.2003.1201821
• [2] Schechner Y Y, Narasimhan S G, Nayar S K.Instant dehazing of images using polarization[C]//Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition.Kauai, HI, USA:IEEE, 2001, 1:I-325-I-332.[DOI:10.1109/CVPR.2001.990493]
• [3] Kopf J, Neubert B, Chen B, et al. Deep photo:model-based photograph enhancement and viewing[J]. ACM Transactions on Graphics , 2008, 27 (5) : #116. DOI:10.1145/1457515.1409069
• [4] Tan R T.Visibility in bad weather from a single image[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Anchorage, Alaska, USA:IEEE, 2008:1-8.[DOI:10.1109/CVPR.2008.4587643]
• [5] Kim J H, Jang W D, Sim J Y, et al. Optimized contrast enhancement for real-time image and video dehazing[J]. Journal of Visual Communication and Image Representation , 2013, 24 (3) : 410–425. DOI:10.1016/j.jvcir.2013.02.004
• [6] Fattal R. Dehazing using color-lines[J]. ACM Transactions on Graphics , 2014, 34 (1) : #13. DOI:10.1145/2651362
• [7] 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. DOI:10.1109/TPAMI.2010.168
• [8] He K M, Sun J, Tang X O. Guided image filtering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2013, 35 (6) : 1397–1409. DOI:10.1109/TPAMI.2012.213
• [9] Gibson K B, Vo D T, Nguyen T Q. An investigation of dehazing effects on image and video coding[J]. IEEE Transactions on Image Processing , 2012, 21 (2) : 662–673. DOI:10.1109/TIP.2011.2166968
• [10] Wang J B, He N, Zhang L L, et al. Single image dehazing witha physical model and dark channel prior[J]. Neurocomputing , 2015, 149 : 718–728. DOI:10.1016/j.neucom.2014.08.005
• [11] Chen Z Y, Jiang T T, Tian Y H.Quality Assessment for Comparing Image Enhancement Algorithms[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Columbus, OH, USA:IEEE, 2014:3003-3010.[DOI:10.1109/CVPR.2014.384]
• [12] Chang C C, Lin C J. LIBSVM:a library for support vector machines[J]. ACM Transactions on Intelligent Systems and Technology , 2011, 2 (3) : #27. DOI:10.1145/1961189.1961199
• [13] Cho H, Lee H, Kang H, et al. Bilateral texture filtering[J]. ACM Transactions on Graphics , 2014, 33 (4) : #128. DOI:10.1145/2601097.2601188