2.765

2022影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于分类学习的去雾后图像质量评价算法

南栋 毕笃彦 马时平 凡遵林 何林远

南栋, 毕笃彦, 马时平, 凡遵林, 何林远. 基于分类学习的去雾后图像质量评价算法. 自动化学报, 2016, 42(2): 270-278. doi: 10.16383/j.aas.2016.c140854
引用本文: 南栋, 毕笃彦, 马时平, 凡遵林, 何林远. 基于分类学习的去雾后图像质量评价算法. 自动化学报, 2016, 42(2): 270-278. doi: 10.16383/j.aas.2016.c140854
NAN Dong, BI Du-Yan, MA Shi-Ping, FAN Zun-Lin, HE Lin-Yuan. A Quality Assessment Method with Classified-learning for Dehazed Images. ACTA AUTOMATICA SINICA, 2016, 42(2): 270-278. doi: 10.16383/j.aas.2016.c140854
Citation: NAN Dong, BI Du-Yan, MA Shi-Ping, FAN Zun-Lin, HE Lin-Yuan. A Quality Assessment Method with Classified-learning for Dehazed Images. ACTA AUTOMATICA SINICA, 2016, 42(2): 270-278. doi: 10.16383/j.aas.2016.c140854

基于分类学习的去雾后图像质量评价算法

doi: 10.16383/j.aas.2016.c140854
基金项目: 

国家自然科学基金 61379104

国家自然科学基金 61372167

详细信息
    作者简介:

    毕笃彦  空军工程大学航空航天工程学院教授.1997年获得法国图尔大学博士学位.主要研究方向为图像处理和模式识别.E-mail:biduyan@126.com

    马时平  空军工程大学航空航天工程学院副教授.2004年获得空军工程大学博士学位.主要研究方向为图像处理和计算机视觉.E-mail:mashiping@163.com

    凡遵林  空军工程大学航空航天工程学院博士研究生.主要研究方向为图像增强和去雾.E-mail:zunlinfan@163.com

    何林远  空军工程大学航空航天工程学院讲师.2008年获得空军工程大学硕士学位.主要研究方向为图像增强和去雾.E-mail:helinyuan@126.com

    通讯作者:

    南栋  空军工程大学航空航天工程学院博士研究生.2011年获得空军工程大学硕士学位.主要研究方向为图像增强和去雾.本文通信作者.E-mail:nd.tian_53@163.com

A Quality Assessment Method with Classified-learning for Dehazed Images

Funds: 

National Natural Science Foundation of China 61379104

National Natural Science Foundation of China 61372167

More Information
    Author Bio:

    Professor at the Institute of Aeronautics and Astronautics, Air Force Engineering University. He received his Ph. D. degree from TUER University, France in 1997. His research interest covers image processing and pattern recognition

    Associate professor at the Institute of Aeronautics and Astronautics, Air Force Engineering University. He received his Ph. D. degree from Air Force Engineering University in 2004. His research interest covers image processing and computer vision

    Ph. D. candidate at the Institute of Aeronautics and Astronautics, Air Force Engineering University. His research interest covers image enhancement and image dehazing

    Lecturer at the Institute of Aeronautics and Astronautics, Air Force Engineering University. He received his master degree from Air Force Engineering University in 2008. His research interest covers image processing and computer vision

    Corresponding author: NAN Dong Ph. D. candidate at the Institute of Aeronautics and Astronautics, Air Force Engineering University. He received his master degree from Air Force Engineering University in 2011. His research interest covers image enhancement and image dehazing. Corresponding author of this paper
  • 摘要: 针对现有去雾后图像质量评价算法少、针对性弱和有效性差等问题, 本文提出一种基于分类学习的去雾后图像质量评价算法.该算法通过分析去雾后图像本身所蕴含的质量特征, 提取出基于图像增强、图像复原、统计先验以及人类视觉系统 (Human visual system, HVS) 的度量指标; 并在本文数据库基础上, 利用支持向量机 (Support vector machine, SVM) 将质量评价问题转换为分类问题.实验结果表明, 该算法与已有评价方法相比, 在获得高效分类评价结果的同时, 具有较好的实用性和主观一致性.
  • 图  1  噪声对去雾前后图像的影响

    Fig.  1  The influences of noise to hazy image and dehazed image

    图  2  基于复原的图像特征描述

    Fig.  2  The image feature descriptor based on restoration

    图  3  噪色度梯度直方图分布示意图[15]

    Fig.  3  Distribution of chromaticity gradient histogram[15]

    图  4  HSV空间统计结果[4]

    Fig.  4  The statistical results of HSV space[4]

    图  5  图像去雾效果示例

    Fig.  5  The results of image dehazing

    图  6  SVM原理

    Fig.  6  The principle of SVM

    图  7  “麦田”在不同 $\alpha$ 取值下的去雾效果

    Fig.  7  The results of image dehazing with different $\alpha$ of "cornfield"

    图  8  “城市”在不同 $\alpha$ 取值下的去雾效果

    Fig.  8  The results of image dehazing with different $\alpha$ of "city"

    表  1  标签矩阵 $L(j)$ 与 $x_j$ 、 $y_j$ 的映射关系

    Table  1  The mapping relationships between $L(j)$ and $x_j$ , $y_j$

    $x_j:y_j$
    -- +1 +1
    -1 -- +1
    -1 -1 --
    下载: 导出CSV

    表  2  纵向实验结果

    Table  2  The results of vertical experiment

    类别 正确率 (%)
    $T_1$ 70.7333
    $T_2$ 65.5667
    $T_3$ 73.8000
    $T_4$ 67.7667
    $T_5$ 62.9333
    $T_6$ 68.9667
    $T_7$ 72.2000
    本文 97.3030
    下载: 导出CSV

    表  3  横向实验结果

    Table  3  The results of horizontal experiment

    类别 正确率 (%) 时间 (s)
    文献[13] 87.0333 71.0256
    文献[4] 89.3000 1.9436
    文献[22] 92.4333 46.2982
    本文 97.3030 3.5429
    下载: 导出CSV

    表  4  图 7去雾后图像质量排序

    Table  4  The quality ranking of dehazed images in Fig. 7

    类别 排序
    参考排序 (e) > (d) > (f) > (c) > (g)
    文献[13] (d) > (e) > (f) > (c) > (g)
    文献[4] (d) > (e) > (c) > (f) > (g)
    文献[22] (e) > (d) > (c) > (f) > (g)
    本文 (e) > (d) > (f) > (c) > (g)
    下载: 导出CSV

    表  5  图 8去雾后图像质量排序

    Table  5  The quality ranking of dehazed images in Fig. 8

    类别 排序
    参考排序 (e) > (f) > (g) > (d) > (c)
    文献[13] (e) > (d) > (f) > (g) > (c)
    文献[4] (e) > (d) > (f) > (g) > (c)
    文献[22] (e) > (f) > (d) > (g) > (c)
    本文 (e) > (f) > (g) > (d) > (c)
    下载: 导出CSV
  • [1] He K M, Sun J, Tang X O. Single image haze removal using dark channel prior. In:Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition. Miami, USA:IEEE, 2009. 1956-1963
    [2] 刘海波, 杨杰, 吴正平, 张庆年, 邓勇.基于暗通道先验和Retinex理论的快速单幅图像去雾方法.自动化学报, 2015, 41(7):1264-1273 http://www.aas.net.cn/CN/abstract/abstract18700.shtml

    Liu Hai-Bo, Yang Jie, Wu Zheng-Pin, Zhang Qing-Nian, Deng Yong. A fast single image dehazing method based on dark channel prior and Retinex theory. Acta Automatica Sinica, 2015, 41(7):1264-1273 http://www.aas.net.cn/CN/abstract/abstract18700.shtml
    [3] 吴迪, 朱青松.图像去雾的最新研究进展.自动化学报, 2015, 42(2):221-239 http://www.aas.net.cn/CN/abstract/abstract18603.shtml

    Wu Di, Zhu Qing-Song. The latest research progress of image dehazing. Acta Automatica Sinica, 2015, 42(2):221-239 http://www.aas.net.cn/CN/abstract/abstract18603.shtml
    [4] 南栋, 毕笃彦, 查宇飞, 张泽, 李权合.基于参数估计的无参考型图像质量评价算法.电子与信息学报, 2013, 35(9):2066-2072 http://www.cnki.com.cn/Article/CJFDTOTAL-DZYX201309006.htm

    Nan Dong, Bi Du-Yan, Zha Yu-Fei, Zhang Ze, Li Quan-He. A no-reference image quality assessment method based on parameter estimation. Journal of Electronics and Information Technology, 2013, 35(9):2066-2072 http://www.cnki.com.cn/Article/CJFDTOTAL-DZYX201309006.htm
    [5] Tan R T. Visibility in bad weather from a single image. In:Proceedings of the 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, USA:IEEE, 2008. 1-8
    [6] Fattal R. Single image dehazing. ACM Transactions on Graphics, 2008, 27(3), doi: 10.1145/1360612.1360671
    [7] Kopf J, Neubert B, Chen B, Cohen M, Cohen-Or D, Deussen O, Uyttendaele M, Lischinski D. Deep photo:model-based photograph enhancement and viewing. ACM Transactions on Graphics, 2008, 27(5), doi: 10.1145/1409060.1409069
    [8] Tarel J P, Hautiére N. Fast visibility restoration from a single color or gray level image. In:Proceedings of the 12th IEEE International Conference on Computer Vision. Kyoto, Japan:IEEE, 2009. 2201-2208
    [9] Hautiére N, Tarel J P, Aubert D, Dumont É. Blind contrast enhancement assessment by gradient ratioing at visible edges. Image Analysis and Stereology Journal, 2008, 27(2):87-95 doi: 10.5566/ias.v27.p87-95
    [10] 郭璠, 蔡自兴.图像去雾算法清晰化效果客观评价方法.自动化学报, 2012, 38(9):1410-1419 doi: 10.3724/SP.J.1004.2012.01410

    Guo Fan, Cai Zi-Xing. Objective assessment method for the clearness effect of image defogging algorithm. Acta Automatica Sinica, 2012, 38(9):1410-1419 doi: 10.3724/SP.J.1004.2012.01410
    [11] 李大鹏, 禹晶, 肖创柏.图像去雾的无参考客观质量评测方法.中国图象图形学报, 2011, 16(9):1753-1757 http://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201109030.htm

    Li Da-Peng, Yu Jing, Xiao Chuang-Bai. No-reference quality assessment method for defogged images. Journal of Image and Graphics, 2011, 16(9):1753-1757 http://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201109030.htm
    [12] Wang Y K, Fan C T. Single image defogging by multiscale depth fusion. IEEE Transactions on Image Processing, 2014, 23(11):4826-4837 doi: 10.1109/TIP.2014.2358076
    [13] Moorthy A K, Bovik A C. Blind image quality assessment:from natural scene statistics to perceptual quality. IEEE Transactions on Image Processing, 2011, 20(12):3350-3364 doi: 10.1109/TIP.2011.2147325
    [14] Gao Y Y, Hu H M, Wang S H, Li B. A fast image dehazing algorithm based on negative correction. Signal Processing, 2014, 130:380-398
    [15] Nan D, Bi D Y, Liu C, Ma S P, He L Y. A Bayesian framework for single image dehazing considering noise. The Scientific World Journal, 2014, 2014:651986
    [16] Liu X H, Tanaka M, Okutomi M. Noise level estimation using weak textured patches of a single noisy image. In:Proceedings of the 19th IEEE International Conference on Image Processing. Orlando, Florida, USA:IEEE, 2012. 665-668
    [17] 邵宇, 孙富春, 刘莹.基于局部结构张量的无参考型图像质量评价方法.电子与信息学报, 2012, 34(8):1779-1785 http://www.cnki.com.cn/Article/CJFDTOTAL-DZYX201208000.htm

    Shao Yu, Sun Fu-Chun, Liu Ying. A no-reference image quality assessment method using local structure tensor. Journal of Electronics and Information Technology, 2012, 34(8):1779-1785 http://www.cnki.com.cn/Article/CJFDTOTAL-DZYX201208000.htm
    [18] Mannos J L, Sakrison D J. The effects of a visual fidelity criterion of the encoding of images. IEEE Transactions on Information Theory, 1974, 20(4):525-536 doi: 10.1109/TIT.1974.1055250
    [19] Tang K T, Yang J C, Wang J. Investigating haze-relevant features in a learning framework for image dehazing. In:Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, USA:IEEE, 2014. 2995-3002
    [20] 张凯军, 梁循.一种改进的显性多核支持向量机.自动化学报, 2014, 40(10):2288-2294 http://www.aas.net.cn/CN/abstract/abstract18503.shtml

    Zhang Kai-Jun, Liang Xun. An improved domain multiple kernel support vector machine. Acta Automatica Sinica, 2014, 40(10):2288-2294 http://www.aas.net.cn/CN/abstract/abstract18503.shtml
    [21] Chang C C, Lin C J. LIBSVM:a library for support vector machines[Online], available:http://www.csie.ntu.tw/cjlin/libsvm, 2007, January 1
    [22] Saad M A, Bovik A C, Charrier C. Blind image quality assessment:a natural scene statistics approach in the DCT domain. IEEE Transactions on Image Processing, 2012, 21(8):3339-3352 doi: 10.1109/TIP.2012.2191563
  • 加载中
图(8) / 表(5)
计量
  • 文章访问数:  3178
  • HTML全文浏览量:  247
  • PDF下载量:  1310
  • 被引次数: 0
出版历程
  • 收稿日期:  2014-12-09
  • 录用日期:  2015-10-23
  • 刊出日期:  2016-02-01

目录

    /

    返回文章
    返回