基于自然场景统计的无参考图像质量评价

doi:10.16182/j.issn1004731x.joss.201612004

系统仿真学报 ›› 2016, Vol. 28 ›› Issue (12): 2903-2911.doi: 10.16182/j.issn1004731x.joss.201612004

基于自然场景统计的无参考图像质量评价

李海洋^1,2, 曹伟国^1,2, 李诗锐^1,2, 陶克路^1,2, 李华^1,2

1.中国科学院计算技术研究所智能信息处理重点实验室,北京 100190;
2.中国科学院大学,北京 100049

收稿日期:2015-03-24 修回日期:2015-05-04 出版日期:2016-12-08 发布日期:2020-08-13
作者简介:Li Haiyang (1986-), male, Beijing, China, Ph.D. student, research area: image processing, 3D reconstruction.
基金资助:
National Natural Science Foundation of China (61227802, 61379082, 61100129)

Blind Image Quality Assessment Based on Natural Scene Statistics

Li Haiyang^1,2, Cao Weiguo^1,2, Li Shirui^1,2, Tao Kelu^1,2, Li Hua^1,2

1. Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences, Beijing, 100190, China;
2. University of Chinese Academy of Sciences, Beijing, 100049, China

Received:2015-03-24 Revised:2015-05-04 Online:2016-12-08 Published:2020-08-13
About author:Li Haiyang (1986-), male, Beijing, China, Ph.D. student, research area: image processing, 3D reconstruction.
Supported by:
National Natural Science Foundation of China (61227802, 61379082, 61100129)

摘要/Abstract

摘要： 目前基于自然场景统计的无参考图像质量评价方法BRISQUE算法(Blind/Referenceless Image Spatial Quality Evaluator)是这类算法的典型代表;但它仅在原始图像基础上做统计分析,且忽略了各特征间的差异性。由此提出了新的改进算法BRISQUEs,并通过三个步骤实现：将被测图像的梯度图做去均值对比归一化处理,在此基础上构造新的特征向量来评价图像质量;将BRISQUE中的关键特征进行适当加权,并对图像再次评价;平均上述两次评价来进一步降低算法的偏差。通过LIVE数据库上的实验,BRISQUEs的统计评价性能明显好于之前的无参考评价算法,也要优于多尺度结构相似度指标。

关键词: 无参考质量评估, 自然场景统计, 梯度图, 关键特征

Abstract: Nowadays blind/referenceless image spatial quality evaluator (BRISQUE) based on natural scene statistics is one of the state-of-the-art no-reference algorithms. But it only analyzes the original image and ignores the difference of the features constructed. Here an improved algorithm BRISQUEs is proposed and implemented by three steps. First, we apply mean subtracted contrast normalized to the gradient images and construct a new feature vector to assess quality. Second, we weight some key features of BRISQUE to improve assessment. After the two assessments obtained, a further average is made to weaken the bias from different assessments. Through the experiments on the LIVE IQA database, our approach has a remarkable performance than previous no-reference algorithms and is statistically superior to the popular multi-scale structural similarity index.

Key words: no-reference image quality assessment, natural scene statistics, gradient image, key feature

中图分类号:

TP391.9

李海洋, 曹伟国, 李诗锐, 陶克路, 李华. 基于自然场景统计的无参考图像质量评价[J]. 系统仿真学报, 2016, 28(12): 2903-2911.

Li Haiyang, Cao Weiguo, Li Shirui, Tao Kelu, Li Hua. Blind Image Quality Assessment Based on Natural Scene Statistics[J]. Journal of System Simulation, 2016, 28(12): 2903-2911.

参考文献

[1] Bovik A C, Wang Z.Modern Image Quality Assessment[M]. New York USA: Morgan and Claypool, 2006.
[2] 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 (S1057-7149), 2004, 13(4): 600-612.
[3] Wang Z, Simoncelli E P, Bovik A C.Multi-scale structural similarity for image quality assessment[C] // Proceedings of Asilomar Conference on Signals, Systems and Computers, 2004. USA: IEEE, 2004: 1398-1402.
[4] Sheikh H R, Bovik A C.Image information and visual quality[J]. IEEE Transactions on Image Processing (S1057-7149), 2006, 15(2): 430-444.
[5] Larson E C, Chandler D M.Most apparent distortion: full-reference image quality assessment and the role of strategy[J]. Journal of Electronic Imaging (S1017-9909), 2010, 19(1): 143-153.
[6] Li Q, Wang Z.Reduced-reference image quality assessment using divisive normalization-based image representation[J]. IEEE Journal of Selected Topics in Signal Processing (S1932-4553), 2009, 3(2): 202-211.
[7] Soundararajan R, Bovik A C.RRED indices: Reduced reference entropic differencing for image quality assessment[J]. IEEE Transactions on Image Processing (S1057-7149), 2011, 21(2): 517-526.
[8] Srivastava A, Lee A B, Simoncelli E P, et al.On advances in statistical modeling of natural images[J]. Journal of Mathematical Imaging and Vision (S0924-9907), 2003, 18(1): 17-33.
[9] Ruderman D L.The statistics of natural images[J]. Network: Computation in Neural Systems (S0954-898X), 1994, 5(4): 517-548.
[10] Moorthy A K, Bovik A C.Blind image quality assessment: From natural scene statistics to perceptual quality[J]. IEEE Transactions on Image Processing (S1057-7149), 2011, 20(12): 3350-3364.
[11] Saad M A, Bovik A C, Charrier C.Blind image quality assessment: A natural scene statistics approach in the DCT domain[J]. IEEE Transactions on Image Processing (S1057-7149), 2012, 21(8): 3339-3352.
[12] Mittal A, Moorthy A K, Bovik A C.No-reference image quality assessment in the spatial domain[J]. IEEE Transactions on Image Processing (S1057-7149), 2012, 21(12): 4695-4708.
[13] Zhang Y, Chandler D M.No-reference image quality assessment based on log-derivative statistics of natural scenes[J]. Journal of Electronic Imaging (S1017-9909), 2013, 22(4): 043025.
[14] Ye P, Doermann David.No-reference image quality assessment using visual codebooks[J]. IEEE Transactions on Image Processing (S1057-7149), 2012, 21(7): 3129-3138.
[15] Ye P, Kumar J, Kang L.Real-time no-reference image quality assessment based on filter learning[C]// CVPR, 2013. USA: IEEE, 2013: 987-994.
[16] Zhang Y, Moorthy A K, Chandler D M, et al.C-DIIVINE: No-reference image quality assessment based on local magnitude and phase statistics of natural scenes[J]. Signal Processing: Image Communication (S0923-5965), 2014, 29(7): 725-747.
[17] Sheikh H R, Wang Z, Cormack L, et al. LIVE Image Quality Assessment Database Release 2 [DB/OL]. (2014-06-19) [2015-03-20], http://live.ece.utexas.edu/ research/ quality
[18] Sharifi K, Leon-Garcia A.Estimation of shape parameter for generalized Gaussian distribution in subband decompositions of video[J]. IEEE Transactions on Circuits Systems Video Technology (S1051-8215), 1995, 5(1): 52-56.
[19] Lasmar N E, Stitou Y, Berthoumieu Y.Multiscale skewed heavy tailed model for texture analysis[C]// Proceedings of the 16th IEEE International Conference on Image Processing. USA: IEEE, 2009: 2281-2284.
[20] Schölkopf B, Smola A J, Williamson R C, et al.New support vector algorithms[J]. Neural Computation (S0899-7667), 2000, 12(5): 1207-1245.
[21] Chang C C, Lin C J. LIBSVM: A Library for Support Vector Machines [DB/OL]. (2014-10-20) [2015-03-20], http://www.csie. ntu.edu.tw/~cjlin/libsvm.
[22] Sheikh H R, Sabir M F, Bovik A C.A statistical evaluation of recent full reference image quality assessment algorithms[J]. IEEE Transactions on Image Processing (S1057-7149), 2006, 15(11): 3441-3452.
[23] Wang Z, Bovik A C, Sheikh E P, et al. Simoncelli. SSIM Code [DB/OL]. (2014-11-09) [2015-03-20], http://www. ece. uwaterloo.ca/~ z70wang/research/ssim.
[24] Wang Z, Simoncelli E P, Bovik A C. MS-SSIM Code [DB/OL]. (2014-11-09) [2015-03-20], https://ece. uwaterloo.ca/~z70wang/research/ iwssim.
[25] Sheikh H R, Bovik A C. VIF Code [DB/OL]. (2005-05-14) [2015-03-20], http://live. ece.utexas.edu/research/Quality / vifvec_release.zip.
[26] Larson E C, Chandler D M. MAD Code [DB/OL]. (2011-10-07) [2015-03-20], http://vision.okstate.edu/mad/ MAD_index_2011_10_07.zip.
[27] Mittal A, Moorthy A K, Bovik A C. BRISQUE Code 2 [DB/OL]. (2012-01-13) [2015-03-20], http://live.ece. utexas. edu/ research/quality/ BRISQUE_release.zip.

基于自然场景统计的无参考图像质量评价

Blind Image Quality Assessment Based on Natural Scene Statistics

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价