可控金字塔分解的立体图像质量评价方法

摘要/Abstract

摘要： 近年来，随着3D电影、3D电视和3D游戏等技术的普及，立体图像处理已经成为研究热点，而对立体图像的质量评价就是此领域中的一项重要技术。本文首先使用最小能量差的方法得到左右视图的视差图，再对左视图、右视图和视差图分别进行4尺度12个方向的可控金字塔（Steerable Pyramid）分解得到3条高频子带114条方向子带，对左右视图分解后相对应的48条方向子带进行二元广义高斯分布拟合，提取其形状参数和尺度参数，并对所有方向子带提取跨尺度相关性、空间相关性等特征信息，最后将这些特征输入支持向量回归（SVR）训练预测得到立体图像质量评分。实验结果表明，该质量评价模型在LIVE 3D数据库上的性能指标SROCC和CC均在0.93以上，与人类的主观评价具有较好的一致性。

关键词: 支持向量回归（SVR）

Abstract: In recent years, with 3D movies, 3D TV and 3D games development, stereo image has become a hot research topic. The quality assessment of stereo image is an important technology in this field. Against this problem, we utilize the steerable pyramid decomposition which has 4 scales and 12 orientations on the left image, right image and the disparity map that gain from the minimum error energy in the left and right image to get 3 high frequency sub-bands and 114 orientation sub-bands. We extract bivariate generalized Gaussian distribution of scale and shape parameters from 48 sub-bands which are got from the left image and the right image. Then we extract correlations across scales, spatial correlation statistical features from all orientation sub-bands. These features put into the support vector regression (SVR) trained to predict the stereo image quality score. The experimental results shows that the objective evaluation model used in this paper, SROCC and CC are more than 0.93 in the 3D LIVE database, and it has good consistency with the subjective evaluation of human.

Key words: Support vector regression (SVR)

中图分类号:

TP391

刘新会桑庆兵. 可控金字塔分解的立体图像质量评价方法[J]. J4, 2017, 34(6): 672-681.

参考文献

[1]Chen M J, Cormack L K, Bovik A C.No-reference quality assessment of natural stereopairs[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2013, 22(9):3379-3391 [2]Xiong Runsheng, Li Chaofeng, Zhang Wei.No-reference Stereoscopic Iillage Quality Assessment Based OR Wavelet Transform[J][J].Computer Science vol, 2015, 42(9):282-284 [3]Moorthy A K, Bovik A C.Blind image quality assessment: from natural scene statistics to perceptual quality[J].IEEE Transactions on Image Processing, 2011, 20(12):3350-3364 [4]Chikkerur S, Sundaram V, Reisslein M, et al.Objective Video Quality Assessment Methods: A Classification,Review,and Performance Comparison[J].IEEE Transactions on Broadcasting, 2011, 57(2):165-182 [5]Moorthy A K, Su C C, Mittal A, et al.Subjective evaluation of stereoscopic image quality[J].Signal Processing Image Communication, 2013, 28(8):870-883 [6]Seuntiens P, Meesters L, Ijsselsteijn W.Perceived quality of compressed stereoscopic images: Effects of symmetric and asymmetric JPEG coding and camera separation[J].Acm Transactions on Applied Perception, 2006, 3(2):95-109 [7] Chen M J, Bovik A C, Cormack L K.Study on distortion conspicuity in stereoscopically viewed 3D images[C]// Ivmsp Workshop, 2011 IEEE. 2011:24-29 [8] Chen M J, Kwon D K, Bovik A C.Study of subject agreement on stereoscopic video quality[C]// Image Analysis and Interpretation. IEEE, 2012:173-176 [9]Zhang Jing, Sang Qingbing.Quality assessment method of color stereoscopic images[J].Journal of Computer Applications, 2015, 35(3):816-820 [10] Mittal A, Moorthy A K, Bovik A C.Blind/Referenceless Image Spatial Quality Evaluator[C]. Circuits, Systems and Computers, 1977. Conference Record. 1977 11th Asilomar Conference on. 2011:723-727. [11] You J, Wang X, Xing L, et al.Perceptual Quality Assessment for Stereoscopic Images Based on 2D Image Quality Metrics and Disparity Analysis[C]// International Workshop on Video Processing and Quality Metrics for Consumer Electronics. 2010. [12] Sazzad Z M P, Akhter R, Baltes J, et al.Objective no-reference stereoscopic image quality prediction based on 2D image features and relative disparity[J].[J].Advances in Multimedia, 2012, :- [13]Bensalma R, Larabi M C.A perceptual metric for stereoscopic image quality assessment based on the binocular energy[J].Multidimensional Systems and Signal Processing, 2013, 24(2):281-316 [14]LIN Y H, WU J L.Quality assessment of stereoscopic 3D image compression by binocular integration behaviors[J].IEEE Transactions on Image Processing, 2014, 23(4):1527-1542 [15]Weijun Tian, Feng Shao, Gangyi Jiang.Blind Image Quality Assessment for Stereoscopic Images via Deep Learning[J][J].Journal of Computer-Aided Design&Computer Graphics, 2016, 28(6):958-976 [16]Hu Gang, Ji Xiaomin, Liu Zhe et al.Regional Feature Self-Adaptive Image Fusion Method Based on Nonsubsampled Steerable Pyramid Transform[J][J].Journal of Computer-Aided Design & Computer Graphics, 2012, 24(1):636-648 [17]Cui Ying, An Image Fusion Algorithm Based on Steerable Pyramid[J.An Image Fusion Algorithm Based on Steerable Pyramid[J][J].Aeronautical Computing Technique, 2011, 41(2):24-27 [18]Freeman W T, Adelson E H.The Design and Use of Steerable Filters[J].IEEE Transactions on Pattern Analysis & Machine Intelligence, 1991, 13(9):891-906 [19]Field D J.Relations between the statistics of natural images and the response properties of cortical cells[J].Journal of the Optical Society of America A Optics & Image Science, 1987, 4(4):2379-94 [20]Sheikh H R, Bovik A C.Image information and visual quality[J].IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2006, 15(2):430-44 [21]Wang Z, 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 [22].Jiang Ming, Research on the method of 3D information acquisition and reconstruction based on binocular visio[D], Kunming: Kunming University of Science and Technology, 2010 [23] Alagoz B B.Obtaining Depth Maps From Color Images By Region Based Stereo Matching Algorithms[J][J]., 2009., :- [24]Su C C, Bovik A C.Bivariate statistical modeling of color and range in natural scenes[J].Proceedings of SPIE - The International Society for Optical Engineering, 2014, 9014(10):2458-2462