去雾图像质量客观评价方法研究

J4 ›› 2016, Vol. 33 ›› Issue (1): 20-28.

去雾图像质量客观评价方法研究

刘海波,汤群芳,杨杰

1 武汉理工大学光纤传感与信号处理教育部重点实验室，湖北武汉 430070；
2 湖南工学院电气与信息工程学院，湖南衡阳 421002；
3 湖南工学院信号与信息处理重点实验室，湖南衡阳 421002

收稿日期:2015-01-29 修回日期:2015-03-11 出版日期:2016-01-28 发布日期:2016-01-06
通讯作者: 杨杰（1960-），女，湖北人，教授，博士生导师，从事信息处理及模式识别方面的研究工作。 E-mail:jieyang509@163.com
作者简介:刘海波（1981-）湖北人，博士生，讲师，从事图像处理及模式识别方面的研究工作。 E-mail:seainlost81@126.com
基金资助:
国家自然科学基金项目（51479159），湖南省教育厅科学研究项目（11C0361，13C206）

Research of objective assessment method for defogged image quality

Liu Haibo, Tang Qunfang, Yang Jie

1 Key Laboratory of Fiber Optic Sensing Technology and Information Processing, Ministry of Education, Wuhan University of Technology, Wuhan 430070,Hubei,China;
2 School of Electrical and Information Engineering, Hunan Institute of Technology, Hengyang 421002,Hunan,China;
3 Key Laboratory of Signal and Information Processing, Hunan Institute of Technology, Hengyang 421002,Hunan,China

Received:2015-01-29 Revised:2015-03-11 Published:2016-01-28 Online:2016-01-06

摘要/Abstract

摘要：

针对目前去雾图像质量客观评价方法少和已有评价方法存在局限性等问题，提出一种去雾图像的客观质量评价新方法。针对去雾图像中存在的Halo效应，采用边缘加权结构相似性测度的方法，将雾化图像和去雾图像的整体轮廓信息与局部纹理细节信息加权来描述图像的结构相似度。从人类视觉感知的角度出发，定义了图像归一化灰度差，用来描述图像的亮度。将彩色图像转换到HSV颜色空间，利用S分量的信息熵和直方图相似度来描述图像的色彩还原能力。最后综合边缘加权结构相似性测度、图像归一化灰度差和色彩加权还原度建立去雾图像质量评价模型，并进行客观评价和比较。实验结果表明，与已有评价方法相比，本文所提评价方法能够获得与主观感受一致的结果，并具有较好的有效性和可靠性。

关键词: 图像处理；图像评价；边缘加权结构相似性测度；归一化灰度差；色彩加权还原度

Abstract:

Since there is lack of methodology to assess the performance of dehazing method and the existing assessment methods have some limitations, an new objective quality assessment method for defogged images is proposed. For the halo artifacts, an edge-weighted structural similarity index is adopted to assign different weights to contour correlation and local texture correlation of the fog image and defogged image, which can represent the structural similarity. From the perspective of human visual perception, the normalized gray difference of defogged image is defined to describe the luminance of the image. Then, the fog image and defogged image are changed from RGB space to HSV space. Information entropy and histogram similarity of S component are used to assess the color performance of defogged image. Finally, with these three factors, a comprehensive image quality assessment model is established, which is used to make the objective evaluation and comparison. Experimental results demonstrate that the proposed method can achieve a consistent result with our subjective perception as well as obtain a good effectiveness and reliability.

Key words: image processing; image assessment; edge-weighted structural similarity index; normalized gray difference; color-weighted reproduction index

中图分类号:

O435.1

刘海波汤群芳杨杰. 去雾图像质量客观评价方法研究[J]. J4, 2016, 33(1): 20-28.

Liu Haibo, Tang Qunfang, Yang Jie. Research of objective assessment method for defogged image quality[J]. J4, 2016, 33(1): 20-28.

参考文献

[1]TAN R.T.. Visibility in Bad Weather from a Single Image [C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, AK, 2008: 1-8.
[2]FATTAL R.Single Image Dehazing[J].ACM Transactions on Graphics, 2008, 27(3):1-9
[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):116:1-10
[4]HE K M, SUN J, TANG X O.Single Image Haze Removal Using Dark Channel Prior[C]. Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Miami, FL, 2009:1956-1963.
[5]TAREL J.P., HAUTIERE N.. Fast Visibility Restoration from a Single Color or Gray Level Image[C]. Proceedings of IEEE International Conference on Computer Vision, Kyoto, 2009:2201-2208.
[6]YU J, LI D P, LIAO Q M.Physics-based Fast Single Image Fog Removal[J].Acta Automatica Sinica, 2011, 37(2):143-149
[7]SUN W, LI D J, LIU H J, et al.Fast single image fog removal based on atmospheric scattering model[J].Optics and Precision Engineering, 2013, 21(4):1040-1046
[8]ZHANG X G, TANG M L, CHEN H, 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
[9]Yao B, Huang L, Liu C P.Research on an objective method to compare the quality of defogged images[C]. Proceeding of the 2009 Chinese Conference on Pattern Recognition, New York, USA: IEEE, 2009:1-5.
[10]Li D P, Yu J, Xiao C B.No-reference quality assessment method for defogged images[J].Journal of Image and Graphics, 2011, 16(9):1753-1757
[11]Guo F, Cai Z X.Objective Assessment Method for the Clearness Effect of Image Defogging Algorithm[J].Acta Automatica Sinica, 2012, 38(9):1410-1419
[12]Zhou W, Bovik A C, Sheikh H R, et al.Image quality assessment: from error visibility to structural similarity[J].IEEE Transactions on Image Processing, 2004, 13(4):600-612
[13]Carnec M, Le Callet P, Barba D.Objective quality assessment of color images based on a generic perceptual reduced reference[J].Image Communication, 2008, 23(4):239-256
[14]Sheikh H R, Bovik A C, Cormack L.No-reference quality assessment using natural scene statistics: JPEG 2000[J].IEEE Transactions on Image Processing, 2005, 14(11):1918-1927
[15]Hautière N.Tarel JP.,Aubert D.,et al.. Blind Contrast Restoration Assessment by Gradient Ratioing at Visible Edges[J].Image Analysis and Stereology, 2008, 27(2):87-95
[16]Fan Y Y, Shen X H, Sang Y J.No reference image sharpness assessment based on contrast sensitivity[J].Optics and Precision Engineering, 2011, 19(10):2485-2493
[17]Lv P, Zhang J Q.Edge-weighted Structural Similarity Index[J].Computer Engineering, 2011, 37(14):226-227
[18]Santiago A F, Raul S J E, Carlos A L, et al.. Image Quality Assessment based on Local Variance[C]. Proceedings of the 28th IEEE EMBS Annual International Conference, New York, USA, 2006:4815-4818.
[19]Xie Z X, Wang Z F, Xiong X L, et al.Color image quality assessment based on noise model of human vision perception and color image quality optimization[J].Journal of Image and Graphics, 2010, 15(10):1454-1464
[20]Nan D, Bi D Y, Zha Y F, et al.A No-reference Image Quality Assessment Method Based on Parameter Estimation[J].Journal of Electronics & Information Technology, 2013, 35(9):2066-2072