半参考和无参考图像质量评价新方法研究
|
摘要
近年来日益增长的图像应用方面的消费需求，促进了人们对客观图像质量评价算法的研究兴趣。客观图像质量评价研究的目标就是开发出能够像人眼一样自动地测量图像质量下降的评价算法，它在图像和视频处理的各个领域中都处于重要地位。根据使用参考图像信息的多少，客观图像质量评价方法可以分为全参考方法、半参考方法和无参考方法。本文主要聚焦半参考和无参考方法，其主要内容总结如下：
(1)提出了一种基于小波变换域的半参考图像质量评价算法，该方法首先对图像进行2尺度小波分解，提取尺度2上的低频小波系数作为图像特征向量，为了度量向量的相似性，本文将图像特征向量看成n维欧氏空间中的一个点，然后分别计算参考图像和失真图像的低频小波系数特征向量在n维欧氏空间中所对应的点之间的欧氏距离D(X,Y)，作为为图像质量评价指标。
(2)提出了一种基于灰度共生矩阵的无参考模糊图像质量评价方法。首先通过LogGabor小波变换生成相位一致图像，然后利用灰度共生矩阵计算相位一致图像的信息熵、能量、对比度、相关性和同质性5个特征，最后利用支持向量回归(Support VectorRegression，SVR)模型学习预测图像质量得分。实验结果表明，该方法能够有效地评价图像的模糊失真。
(3)基于离散余弦变换系数提出了一种无参考模糊图像质量评价方法。该方法首先通过对图像进行离散余弦变换，得到图像的离散余弦变换系数作为图像质量变化的特征向量，然后利用广义回归神经网络模型对此特征向量进行训练学习，预测得到无参考模糊图像质量得分。
(4)提出了一种基于奇异值分解的无参考模糊和噪声图像质量评价方法。该方法首先通过对待评价模糊图像和噪声图像进行高斯低通滤波生成再模糊图像，然后分别对待评价图像和再模糊图像分别进行奇异值分解，得到各自奇异值向量，最后构造奇异值的改变量来作为无参考模糊和噪声图像的质量评价指标。
(5)提出了两种基于奇异值分解的通用无参考图像质量评价方法。第一种方法是基于奇异值的改变量构造了一种通用无参考图像质量评价指标，奇异值的改变量能够反映图像的蚀变情况。第二种提出了基于图像奇异值倒数曲线的通用无参考图像质量评价方法。图像的奇异值倒数曲线近似幂函数曲线，且随着图像失真程度的不同，奇异值倒数曲线的弯曲程度也不相同。根据奇异值倒数曲线的这一特征，本文从面积和曲率两个角度，构造了两种通用无参考图像质量评价指标。
In recent years, the increasing number of demanding consumer image applications hasboosted interest in objective image quality assessment (IQA) algorithms. Objective imagequality assessment aims to automatically measure the quality degradation perceived by thehuman eyes. It is of fundamental importance to address a wide variety of problems in imageand video processing. Based on the availability of the information about the reference image,IQA models can be classified into full-reference (FR), reduced-reference (RR) and noreference (NR) IQA methods. This dissertation focuses on RR-IQA AND NR-IQA, the majorcontents are as follows in general：
First, we propose a novel metric for RRIQA based on wavelet transform. We dothe wavelet decomposition of2scales to images, and extract the low frequency waveletcoefficients of the second scale as the image feature vector. In order to measure thesimilarity of vectors, we see the feature vector as a point in n-dimensional Euclidean space,and calculate the distance between referenc image feature vector and distortion image featurevector in n-dimensional Euclidean space. The distance is regarded as the metric of imagequality.
Second, we propose a no reference blur image quality assessment method based on graylevel co-occurrence matrix extraction phase congruency image feature and supportvector regression (SVR). The method is composed of three steps. First, we use Log Gaborwavelet to generate phase congruency map of the image. Then we calculate the PhaseCongruency map’s features which are entropy, energy, contrast, correlation and homogeneityby gray level co-occurrence matrix. Finally, we predict no-reference blur image quality scoreby using SVR model training and learning.
Third, we propose a no reference blur image quality assessment method based ondiscrete cosine transform. First of all, we do discrete cosine transform to images and extractdiscrete cosine transform coefficients as feature vector, and then used the generalizedregression neural network model to train feature vector to predict image quality. In the threepublic databases, the experimental results show that this method has a good correlation withthe subjective quality score.
Fourth, we propose a new blind blur and noise index for still images using Gaussian blurand Singular Value Decomposition (SVD). The algorithm is composed of three steps. Firstly,a re-blurred image is produced by using Gaussian blur to the test image. Then the singularvalue decomposition is performed to the test image and re-blurred image. Finally, a blur andnoise index is constructed by using the change of singular values. Experimental resultsobtained on four simulated databases show that the proposed algorithm has high correlationwith human judgments when assessing blur distortion of images.
Finally, we propose two universal blind image quality assessment methods. The first is auniversal blind image quality assessment method based on the change of singular value. Thechange of singular value can reflect the distortion of an image. The second is a universal blindimage quality assessment method using a reciprocal singular value curve. The reciprocalsingular value curves of natural images resemble inverse power functions. The bending degree of the reciprocal singular value curve is varies with distortion type and severity. Weconstructed two new general blind IQA indices utilizing the area and curvature of imagereciprocal singular value curves. These two methods do not require prior knowledge of anyimage or distortion, and hence do not require any process of training, hence are "completelyblind" IQA models.
(1)提出了一种基于小波变换域的半参考图像质量评价算法，该方法首先对图像进行2尺度小波分解，提取尺度2上的低频小波系数作为图像特征向量，为了度量向量的相似性，本文将图像特征向量看成n维欧氏空间中的一个点，然后分别计算参考图像和失真图像的低频小波系数特征向量在n维欧氏空间中所对应的点之间的欧氏距离D(X,Y)，作为为图像质量评价指标。
(2)提出了一种基于灰度共生矩阵的无参考模糊图像质量评价方法。首先通过LogGabor小波变换生成相位一致图像，然后利用灰度共生矩阵计算相位一致图像的信息熵、能量、对比度、相关性和同质性5个特征，最后利用支持向量回归(Support VectorRegression，SVR)模型学习预测图像质量得分。实验结果表明，该方法能够有效地评价图像的模糊失真。
(3)基于离散余弦变换系数提出了一种无参考模糊图像质量评价方法。该方法首先通过对图像进行离散余弦变换，得到图像的离散余弦变换系数作为图像质量变化的特征向量，然后利用广义回归神经网络模型对此特征向量进行训练学习，预测得到无参考模糊图像质量得分。
(4)提出了一种基于奇异值分解的无参考模糊和噪声图像质量评价方法。该方法首先通过对待评价模糊图像和噪声图像进行高斯低通滤波生成再模糊图像，然后分别对待评价图像和再模糊图像分别进行奇异值分解，得到各自奇异值向量，最后构造奇异值的改变量来作为无参考模糊和噪声图像的质量评价指标。
(5)提出了两种基于奇异值分解的通用无参考图像质量评价方法。第一种方法是基于奇异值的改变量构造了一种通用无参考图像质量评价指标，奇异值的改变量能够反映图像的蚀变情况。第二种提出了基于图像奇异值倒数曲线的通用无参考图像质量评价方法。图像的奇异值倒数曲线近似幂函数曲线，且随着图像失真程度的不同，奇异值倒数曲线的弯曲程度也不相同。根据奇异值倒数曲线的这一特征，本文从面积和曲率两个角度，构造了两种通用无参考图像质量评价指标。
In recent years, the increasing number of demanding consumer image applications hasboosted interest in objective image quality assessment (IQA) algorithms. Objective imagequality assessment aims to automatically measure the quality degradation perceived by thehuman eyes. It is of fundamental importance to address a wide variety of problems in imageand video processing. Based on the availability of the information about the reference image,IQA models can be classified into full-reference (FR), reduced-reference (RR) and noreference (NR) IQA methods. This dissertation focuses on RR-IQA AND NR-IQA, the majorcontents are as follows in general：
First, we propose a novel metric for RRIQA based on wavelet transform. We dothe wavelet decomposition of2scales to images, and extract the low frequency waveletcoefficients of the second scale as the image feature vector. In order to measure thesimilarity of vectors, we see the feature vector as a point in n-dimensional Euclidean space,and calculate the distance between referenc image feature vector and distortion image featurevector in n-dimensional Euclidean space. The distance is regarded as the metric of imagequality.
Second, we propose a no reference blur image quality assessment method based on graylevel co-occurrence matrix extraction phase congruency image feature and supportvector regression (SVR). The method is composed of three steps. First, we use Log Gaborwavelet to generate phase congruency map of the image. Then we calculate the PhaseCongruency map’s features which are entropy, energy, contrast, correlation and homogeneityby gray level co-occurrence matrix. Finally, we predict no-reference blur image quality scoreby using SVR model training and learning.
Third, we propose a no reference blur image quality assessment method based ondiscrete cosine transform. First of all, we do discrete cosine transform to images and extractdiscrete cosine transform coefficients as feature vector, and then used the generalizedregression neural network model to train feature vector to predict image quality. In the threepublic databases, the experimental results show that this method has a good correlation withthe subjective quality score.
Fourth, we propose a new blind blur and noise index for still images using Gaussian blurand Singular Value Decomposition (SVD). The algorithm is composed of three steps. Firstly,a re-blurred image is produced by using Gaussian blur to the test image. Then the singularvalue decomposition is performed to the test image and re-blurred image. Finally, a blur andnoise index is constructed by using the change of singular values. Experimental resultsobtained on four simulated databases show that the proposed algorithm has high correlationwith human judgments when assessing blur distortion of images.
Finally, we propose two universal blind image quality assessment methods. The first is auniversal blind image quality assessment method based on the change of singular value. Thechange of singular value can reflect the distortion of an image. The second is a universal blindimage quality assessment method using a reciprocal singular value curve. The reciprocalsingular value curves of natural images resemble inverse power functions. The bending degree of the reciprocal singular value curve is varies with distortion type and severity. Weconstructed two new general blind IQA indices utilizing the area and curvature of imagereciprocal singular value curves. These two methods do not require prior knowledge of anyimage or distortion, and hence do not require any process of training, hence are "completelyblind" IQA models.
引文
[1] Wang Z, Bovik A C. Modern Image Quality Assessment [M]. New York: Morgan andClaypool Publishers,2006.13-27.
[2] Nachlieli H, Shaked D. Measuring the quality of quality measures [J]. Image Processing,IEEE Transactions on,2011,20(1):76-87.
[3] Karunasekera S A and Kingsbury N G. A distortion measure for image artifacts based onhuman visual sensitivity [J]. Proc. IEEE Int. Conf. Acoust.,Speech, and Signal Processing,1994,5(4):117–120.
[4] Liu H, Heynderickx I. Visual attention in objective image quality assessment: based oneye-tracking data [J]. Circuits and Systems for Video Technology, IEEE Transactions on,2011,21(7):971-982.
[5] Wang Z. Applications of objective image quality assessment methods [J]. IEEE SignalProcessing Magazine,2011,28(11):123-129.
[6] Wang Z. Demo images and free software for a universal image qualityindex [DB/OL].http://anchovy.ece.utexas.edu/zwang/research/quality/index/demo.html
[7] Wang Z and Bovik A C. Mean squared error：Love it or leave it?-a new look at signalfidelity measures [J]. IEEE Signal Processing Magazine,2009,26(1):98-117.
[8] Rehman A, Rostami M, Wang Z, et al. SSIM-inspired image restoration using sparserepresentation[J]. EURASIP Journal on Advances in Signal Processing,2012,1:1-12.
[9] Ye P, Doermann D. No-reference image quality assessment using visual codebooks [J].Image Processing, IEEE Transactions on,2012,21(7):3129-3138.
[10] Wang Z, Bovik A C. Reduced-and no-reference image quality assessment [J]. SignalProcessing Magazine, IEEE,2011,28(6):29-40.
[11] Kim D O, Park R H. New image quality metric using random projection [J]. ImageProcessing, IET,2012,6(9):1246-1255.
[12] Wang Z, Bovik A C, Sheikh H R, et al. Image quality assessment: from error visibility tostructural similarity [J]. IEEE Trans. Image Processing,2004,13(4):600-612.
[13]郑江云,江巨浪.基于小波第二级系数误差的图像质量评价模型[J].电子学报，2012,40(3):559-563.
[14]杨春玲,高文瑞.基于结构相似的小波域图像质量评价方法的研究[J].电子学报，2009,37(4):845-849.
[15]路文,高新波.一种基于WBCT的自然图像质量评价方法[J].电子学报，2008,36(2):303-308.
[16] Gao Xinbo, Lu Wen, Tao Dacheng, et al. Image Quality Assessment Based on MultiscaleGeometric Analysis [J]. IEEE Transactions On Image Processing,2009,18(7):1409-1423.
[17] Soundararajan R，Bovik A C. RRED Indices：Reduced Reference Entropic Differencingfor Image Quality Assessment [J]. IEEE Transactions on Image Processing,2012,21(2):517-526.
[18] Sheikh H R, Bovik A C, de Veciana G. An information fidelity criterion for image qualityassessment using natural scene statistics [J]. IEEE Trans. Image Processing,2005,14(12):2117-2128.
[19] Sheikh H R, Bovik A C. Image information and visual quality [J]. IEEE Trans. ImageProcessing,2006,15(2):430-444.
[20] Chandler D M, Hemami S S. A wavelet-based visual signal-to-noise ratio for naturalimages [J]. IEEE Trans. Image Processing,2007,16(9):2284-2298.
[21] Moorthy A K, Bovik A C. Perceptually significant spatial pooling techniques for imagequality assessment [J]. In Proc. SPIE Human Vision and Electronic Imaging XIV,2009,7240:724012-724012-11.
[22] Zhang Lin, Zhang Lei, Mou Xuanqin and Zhang David. FSIM: A Feature SimilarityIndex for Image Quality Assessment [J]. IEEE Transactions on Image Processing,2012,20(8):2378-2385.
[23] Rehman A, Wang Z. Reduced-reference image quality assessment by structural similarityestimation [J]. IEEE Transactions on Image Processing,2012,21(8):3378-3389.
[24] Wang Z, Simoncelli E P, Bovik A C. Multiscale structural similarity for image qualityassessment[C]. In: Signals, Systems and Computers,2003. Conference Record of theThirty-Seventh Asilomar Conference on. IEEE,2003,2:1398-1402.
[25] Wang Z, Shang X. Spatial pooling strategies for perceptual image quality assessment[C].In:Image Processing,2006IEEE International Conference on. IEEE,2006:2945-2948.
[26] Bovik E A. Handbook of Image and Video Processing [M]. Second Edition. ElsevierAcademic Press,2002
[27] Wang Z, Simoncelli E P. Reduced-reference image quality assessment using awavelet-domain natural image statistic model [J]. In Proc. SPIE Human Vision andElectronic Imaging X,2005,5666(1):149-159
[28] Gao X B, Lu W, Li X L, et al. Image quality assessment based on multiscale geometricanalysis [J]. IEEE Trans. Image Processing,2009,18(7):1608-1622
[29] Lu W, Gao X, Li X, et al. An image quality assessment metric based contourlet[C].In:Image Processing,2008. ICIP2008.15th IEEE International Conference on. IEEE,2008:1172-1175.
[30]路文,高新波,王体胜.一种基于WBCT的自然图像质量评价方法[J].电子学报,2008,36(2):303-308
[31] Zeng K, Wang Z. Polyview Fusion: A Strategy to Enhance Video-Denoising Algorithms[J]. Image Processing, IEEE Transactions on,2012,21(4):2324-2328.
[32] Yim C, Bovik A C. Quality assessment of deblocked images [J]. Image Processing, IEEETransactions on,2011,20(1):88-98.
[33] Kim S B, Wang Z, Hiremath B. A Bayesian approach for the alignment of high-resolutionNMR spectra [J]. Annals of Operations Research,2010,174(1):19-32.
[34] Ferzli Rony, Karam L J. A No-Reference Objective Image Sharpness Metric Based on theNotion of Just Noticeable Blur (JNB)[J]. IEEE Transactions on Image Processing,2009,18(4):717-728
[35] Pons A, Malo J, Artigas M J and Capilla. Image quality metric based onmultidimensional contrast perception models [J]. Displays,1999,20:93–110.
[36] Wee C Y, Paramesran R, Mukundan R, et al. Image quality assessment by discreteorthogonal moments [J]. Pattern Recognition,2010,43(12):4055-4068.
[37] He L, Gao X, Lu W, et al. Image quality assessment based on S-CIELAB model [J].Signal, Image and Video Processing,2011,5(3):283-290.
[38] Capodiferro L, Jacovitti G, Di Claudio E D. Two-Dimensional Approach toFull-Reference Image Quality Assessment Based on Positional Structural Information [J].Image Processing, IEEE Transactions on,2012,21(2):505-516.
[39] He L, Tao D, Li X, et al. Sparse representation for blind image quality assessment[C]. In:2012IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2012:1146-1153.
[40] Sheikh H R, Bovik A C. A visual information fidelity approach to video qualityassessment[C]. In:The First International Workshop on Video Processing and QualityMetrics for Consumer Electronics.2005:23-25.
[41] Narvekar N D, Karam L J. A no-reference image Blur metric based on the cumulativeprobability of Blur detection (CPBD)[J]. IEEE Transactions on Image Processing,2011,20(9):2678-2683
[42] Li C F, Yuan W, Bovik A C and Wu X. No-reference Blur index using Blur comparisons[J].Electronics Letters,2011,47(17):962-963.
[43] Wang Z, Sheikh H R, Bovik A C. No-reference perceptual quality assessment of JPEGcompressed images[C].In: International Conference on Image Processing.2,2002,1:I-477-I-480.
[44] Narwaria M, Lin W, McLoughlin I V, et al. Fourier transform-based scalable imagequality measure [J]. Image Processing, IEEE Transactions on,2012,21(8):3364-3377.
[45] Liu S, Bovik A C. Efficient DCT-domain blind measurement and reduction of blockingartifacts. IEEE Trans [J]. Circuits and System for Video Technology,2002,12(12):1139-1149
[46] Watson A B. DCTune: A technique for visual optimization of DCT quantizationmatrices for individual images[C]. In:Sid International Symposium Digest of TechnicalPapers. SOCIETY FOR INFORMATION DISPLAY,1993,24:946-946.
[47] Sheikh H R, Bovik A C, and Cormack L. No-reference quality assessmentusing naturalscene statistics: JPEG2000[J]. IEEE Trans. Image Processing,2005,14(11):1918-1927
[48] Wang Z and Bovik A C. Bitplane-by-bitplane shift (BbBShift)-A suggestion for JPEG2000region of interest coding [J].IEEE Signal Processing Letters,2002,9(5):60–162
[49] Marziliano P, Dufaux F, Winkler S and Ebrahimi T. Perceptual blur and ringing metrics:Application to JPEG2000[J]. Signal Processing: Image Communication,2004,19(2):163–172
[50] Li C, Bovik A C and Wu X. Blind image quality assessment using a general regressionneural network [J]. IEEE Trans. Neural Netw,2011,22(5):793–799
[51] Moorthy A K and Bovik A C. Blind image quality assessment: from scene statistics toperceptual quality [J]. IEEE Trans. Image Process,2011,20(12):3350–3364
[52] Saad M A, Bovik A C and Charrier C. Model-based blind image quality assessment: anatural scene statistics approach in the DCT domain [J]. IEEE Trans. Image Process,2012,21(8):3339–3352
[53] Moorthy A K and Bovik A C. A two-step framework for constructing blind image qualityindices [J]. IEEE Signal Process. Lett,201017(5):513–516
[54] Mittal A, Muralidhar G S and Bovik A C. Blind image quality assessment without humantraining using latent quality factors [J]. IEEE Signal Process. Lett,2012,19(2):75–78
[55] Mittal A, Muralidhar G S and Bovik A C. Making a ‘completely blind’ image qualityanalyzer [J]. IEEE Signal Process. Lett,2013,20(3):209–212
[56] Li Chaofeng, Ju Yiwen, Bovik A C, et al. No-training, no-reference image quality indexusing perceptual features [J]. Optical Engineering,2013
[57] Xue W, Zhang L, Mou X. Learning without Human Scores for Blind Image QualityAssessment [C]. In: IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2013.995-1001.
[58] Video Quality Experts Group [DB/OL]. http://www.vqeg.org/
[59] Sheikh H R, Sabir M F and Bovik A C. A statistical evaluation of recent full referenceimage quality assessment algorithms [J]. IEEE Trans. Image Processing,2006,15(11):3440–3451
[60] Serir A, Beghdadi A, Kerouh F. No-Reference Blur Image Quality Measure Based onMultiplicative Mutiresolution Decomposition [J]. Journal of Visual Communication andImage Representation,2013,24(7):911-925.
[61] Larson E C and Chandler D M. Categorical Image Quality (CSIQ) database2009
[DB/OL]. http://vision.okstate.edu/csiq
[62] Sheikh H R, Seshadrinathan K, et al.. Image and Video Quality Assessment Researce atLIVE2004[DB/OL]. http://live.ece.utexas.edu/research/quality
[63] Ponomarenko N, Lukin V, Zelensky A, et al.. TID2008—A database for evaluation offull-reference visual quality assessment metrics [J]. Adv. Modern Radioelectron,2009,10:30–45
[64] Ninassi A, Le Callet P, and Autrusseau F. Subjective Quality Assessment-IVC Database2005[DB/OL]. http://www2.irccyn.ecnantes.fr/ivcdb
[65] Horita Y, Shibata K, Kawayoke Y, and Sazzad Z M P. MICT Image Quality EvaluationDatabase2000[DB/OL]. http://mict.eng.u-toyama.ac.jp/mict/index2.html
[66] BID-Blurred Image Database[DB/OL]http://www.lps.ufrj.br/profs/eduardo/I-mageDatabase.htm
[67] Wang Z, Simoncelli E P. Reduced-reference image quality assessment using awavelet-domain natural image statistic model[C]. In:Proc. of SPIE Human Vision andElectronic Imaging.2005,5666:149-159.
[68] Gao X, Lu W, Tao D and Li X. Image quality assessment based on multiscale geometricanalysis [J]. IEEE Transactions on Image Processing,2009,18(7):1409–1423.
[69] Ma Lin, Lin Weisi, Deng Chenwei, et al. Image retargeting quality assessment： a studyof subjective scores and objective metrics[J]. IEEE Journal of Selected Topics In SignalProcessing,2012,6(6):626–639
[70]叶盛楠,苏开娜等.基于结构信息提取的图像质量评价[J].电子学报，2008,36(5):856-861
[71]杨春玲,高文瑞.基于结构相似的小波域图像质量评价方法的研究[J].电子学报，2009,37(4):845-849
[72]路文,高新波等.一种基于WBCT的自然图像质量评价方法[J].电子学报，2008,36(2):303-308
[73] Hassen R, Wang Z, Salama M. No-reference image sharpness assessment based on localphase coherence measurement[C]. In: IEEE International Conference on Acoustics Speechand Signal Processing (ICASSP),2010:2434-2437.
[74] Chen M J and Bovik A C. No-reference image blur assessment using multiscale gradient[J]. EURASIP Journal on Image and Video Processing,2011,3:1-11
[75] Ciancio A, Da Costa A L N T, et al. No-reference blur assessment of digital picturesbased on multi-feature classifiers [J]. IEEE Trans. Image Processing,2011,20(1):64-75
[76] Saad M A, Bovik A C and Charrier C. A DCT statistics-based blind image quality index[J]. IEEE Signal Process. Lett.,2010,17(6):583–586
[77] Wang S, Rehman A, Wang Z, et al. Ssim-motivated rate-distortion optimization for videocoding[J]. IEEE Transactions on Circuits and Systems for Video Technology,2012,22(4):516-529.
[78] Watson A B, Yang G Y, Solomon J A and Villasenor J. Visibility of wavelet quantizationnoise [J]. IEEE Trans. Image Processing,1997,6(8):1164–1175
[79] Chandler D M, Hemami S S. Additivity models for suprathreshold distortion in quantizedwavelet-coded images[C]. In: International Society for Optics and Photonics ElectronicImaging,2002:105-118.
[80] Selesnick I W, Baraniuk R G, Kingsbury N C. The dual-tree complex wavelettransform[J]. IEEE Signal Processing Magazine,2005,22(6):123-151.
[81] Portilla J, Wainwright V S M J and Simoncelli E P. Image denoising using scale mixturesof gaussians in the wavelet domain [J]. IEEE Trans. Signal Processing,2003,12(11):1331–1338
[82] Brunet D, Vrscay E R, Wang Z. On the mathematical properties of the structuralsimilarity index[J]. Image Processing, IEEE Transactions on,2012,21(4):1488-1499.
[83] Brunet D, Vass J, Vrscay E R, et al. Geodesics of the Structural Similarity index[J].Applied Mathematics Letters,2012,25(11):1921-1925.
[84] Wang Z, Li Q. Information content weighting for perceptual image quality assessment[J].Image Processing, IEEE Transactions on,2011,20(5):1185-1198.
[85] Rostami M, Michailovich O, Wang Z. Image deblurring using derivative compressedsensing for optical imaging application [J]. IEEE Transactions on Image Processing,2012,21(7):3139-3149.
[86] Nikvand N, Wang Z. Image distortion analysis based on normalized perceptualinformation distance [J]. Signal, Image and Video Processing, special issue on HumanVision and Information Theory,2013,7(3):403-410.
[87] Morrone M C and Owens R A. Feature detection from local energy. Pattern RecognitionLetters,1987,6(5)：303–313
[88] Mittal A, Moorthy A K and Bovik A C. No-Reference Image Quality Assessment in theSpatial Domain [J]. IEEE Trans. Image Process.,2012,21(12):4695–4708.
[89] Kovesi P. Image features from phase congruency. Computer Vision Research,1999,1(3):1-26
[90] Peter Kovesi homepage[EB/OL]. http://www.csse.uwa.edu.au/~pk/Research/MatlabFns/index.html
[91] Wang Z and Simoncelli E P. Local phase coherence and the perception of blur [J]. In Adv.Neural Information Processing Systems (NIPS03),2004,16:(5)
[92] Oppenheim A V, Lim J S. The importance of phase in signals [J]. Proceedings of theIEEE,1981,69(5):529-541.
[93] Morrone M C and Burr D C. Feature detection in human vision: A phase dependentenergy model [J]. Proc. R. Soc. Lond. Biological Sciences,1988,23(12):221–245.
[94] Kovesi P. Phase congruency: A low-level image invariant [J]. Psych.Research,2000,64:136–148
[95] Wang Z. Applications of objective image quality assessment methods [J]. IEEE SignalProcessing Magazine,2011,28(6):137-142.
[96] Hassen R, Wang Z, Salama M. Image Sharpness Assessment Based on Local PhaseCoherence[J] IEEE Transactions on Image Processing,2013,22(7):2789-2810.
[97] Haralick R M, Shanmugam K and Dinstein I. Textural Features for Image Classification[J]. IEEE Transactions on Systems, Man, and Cybernetics1973, SMC-3(6):610–621
[98] Hralick R M. Statistical and Structural Approaches toTexture[J]. Proceedings of IEEE,1979,67(5):786-805
[99] Vapnik V. The Nature of Statistical Learning Theory [M]. Berlin, Germany: SpringerVerlag,2000
[100] Narwaria M and Lin W. Objective image quality assessment based on support vectorregression[J]. IEEE Trans. Neural Netw.,2010,21(3):515–519
[101] Yeganeh H and Wang Z. Objective quality assessment of tone mapped images [J]. IEEETransactions on Image Processing,2013,22(2):657-667.
[102] Zhao T, Kwong S, Wang H, et al. Multiview Coding Mode Decision With HybridOptimal Stopping Model [J]. IEEE Transactions on Image Processing,2013,22(4):1598-1609.
[103] Wang S, Rehman A, Wang Z, et al. Perceptual Video Coding Based on SSIM-InspiredDivisive Normalization [J]. IEEE Transactions on Image Processing,2013,22(4):1418-1429.
[104] Varghese G, Wang Z. Video denoising based on a spatiotemporal Gaussian scalemixture model[J]. IEEE Transactions on Circuits and Systems for Video Technology,2010,20(7):1032-1040.
[105] Lyu S, Simoncelli E P. Statistical modeling of images with fields of Gaussian scalemixtures[C]. In: Advances in Neural Information Processing Systems.2006:945-952.
[106] Ye P, Kumar J, Kang L, et al. Unsupervised feature learning framework for no-referenceimage quality assessment [C]. In:2012IEEE Conference on Computer Vision and PatternRecognition (CVPR),2012:1098-1105.
[107] Chih-Jen Lin’s Home Page [EB/OL]. http://www.cise.ntu.edu.tw/~cjlin/libsvm
[108] Bovik A C. The Essential Guide to Image Processing [M]. New York：Academic Press,2009,7:421-426.
[109] Specht D F. A general regression neural network [J]. IEEE Trans. Neural Netw.,1991,2(6):568-576
[110] Chartier S, Boukadoum M and Amiri M. BAM learning of nonlinearly separable tasksby using an asymmetrical output function and reinforcement learning [J]. IEEE Trans.Neural Netw.,2009,20(8):1281–1292.
[111] Tang H, Joshi N, Kapoor A. Learning a blind measure of perceptual image quality [C].In:2011IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2011:305-312.
[112] Chetouani A, Beghdadi A, Chen S and Mostafaoui G. A novel free reference imagequality metric using neural network approach [J]. In Proc.Int. Workshop Video Process.Qual. Metrics Cons. Electron., Scottsdale,AZ,2010(1):1–4.
[113] Li Q, Meng Q, Cai J, Yoshino H and Mochida A. Predicting hourly cooling load in thebuilding: A comparison of support vector machine and different artificial neural networks[J]. Ener. Conv. Manage.,2009,50(1):90–96
[114] Aleksandr S, Alexander G, and Ahmet M E. An SVD-Based Grayscale Image QualityMeasure for Local and Global Assessment [J]. IEEE Trans. Image Process.,2006,15(2):422-429
[115] Narwaria M, Lin W. Objective image quality assessment with Singular ValueDecomposition[C]. In: Proc.5th Int. Workshop Video Processing and Quality Metrics forConsumer Electronics.2010.
[116]楼斌，沈海斌，赵武锋等．基于自然图像统计的无参考图像质量评价[J]．浙江大学学报：工学版，2010,44(2):248-252.
[117]金波，李朝锋，吴小俊．结合NSS和小波变换的无参考图像质量评价[J].中国图象图形学报，2012,17(1):33-39
[118] Cheng C, Wang H. Quality assessment for color images with tucker decomposition[C].In:201219th IEEE International Conference on Image Processing (ICIP),2012:1489-1492.
[119] Abdi H,Williams L J. Principal component analysis.Wiley Interdisciplinary Reviews:Computational Statistics,2010,2:433–459
[120] Andrews H C, Patterson C L. Singular Value Decomposition Image Coding [J]. IEEETrans. on Communications,1976(4):425-432.
[121] Pyatykh S, Hesser J, Zheng L. Image noise level estimation by principal componentanalysis [J]. IEEE Trans. Image Process.,2012,22(12):5226-5237.
[2] Nachlieli H, Shaked D. Measuring the quality of quality measures [J]. Image Processing,IEEE Transactions on,2011,20(1):76-87.
[3] Karunasekera S A and Kingsbury N G. A distortion measure for image artifacts based onhuman visual sensitivity [J]. Proc. IEEE Int. Conf. Acoust.,Speech, and Signal Processing,1994,5(4):117–120.
[4] Liu H, Heynderickx I. Visual attention in objective image quality assessment: based oneye-tracking data [J]. Circuits and Systems for Video Technology, IEEE Transactions on,2011,21(7):971-982.
[5] Wang Z. Applications of objective image quality assessment methods [J]. IEEE SignalProcessing Magazine,2011,28(11):123-129.
[6] Wang Z. Demo images and free software for a universal image qualityindex [DB/OL].http://anchovy.ece.utexas.edu/zwang/research/quality/index/demo.html
[7] Wang Z and Bovik A C. Mean squared error：Love it or leave it?-a new look at signalfidelity measures [J]. IEEE Signal Processing Magazine,2009,26(1):98-117.
[8] Rehman A, Rostami M, Wang Z, et al. SSIM-inspired image restoration using sparserepresentation[J]. EURASIP Journal on Advances in Signal Processing,2012,1:1-12.
[9] Ye P, Doermann D. No-reference image quality assessment using visual codebooks [J].Image Processing, IEEE Transactions on,2012,21(7):3129-3138.
[10] Wang Z, Bovik A C. Reduced-and no-reference image quality assessment [J]. SignalProcessing Magazine, IEEE,2011,28(6):29-40.
[11] Kim D O, Park R H. New image quality metric using random projection [J]. ImageProcessing, IET,2012,6(9):1246-1255.
[12] Wang Z, Bovik A C, Sheikh H R, et al. Image quality assessment: from error visibility tostructural similarity [J]. IEEE Trans. Image Processing,2004,13(4):600-612.
[13]郑江云,江巨浪.基于小波第二级系数误差的图像质量评价模型[J].电子学报，2012,40(3):559-563.
[14]杨春玲,高文瑞.基于结构相似的小波域图像质量评价方法的研究[J].电子学报，2009,37(4):845-849.
[15]路文,高新波.一种基于WBCT的自然图像质量评价方法[J].电子学报，2008,36(2):303-308.
[16] Gao Xinbo, Lu Wen, Tao Dacheng, et al. Image Quality Assessment Based on MultiscaleGeometric Analysis [J]. IEEE Transactions On Image Processing,2009,18(7):1409-1423.
[17] Soundararajan R，Bovik A C. RRED Indices：Reduced Reference Entropic Differencingfor Image Quality Assessment [J]. IEEE Transactions on Image Processing,2012,21(2):517-526.
[18] Sheikh H R, Bovik A C, de Veciana G. An information fidelity criterion for image qualityassessment using natural scene statistics [J]. IEEE Trans. Image Processing,2005,14(12):2117-2128.
[19] Sheikh H R, Bovik A C. Image information and visual quality [J]. IEEE Trans. ImageProcessing,2006,15(2):430-444.
[20] Chandler D M, Hemami S S. A wavelet-based visual signal-to-noise ratio for naturalimages [J]. IEEE Trans. Image Processing,2007,16(9):2284-2298.
[21] Moorthy A K, Bovik A C. Perceptually significant spatial pooling techniques for imagequality assessment [J]. In Proc. SPIE Human Vision and Electronic Imaging XIV,2009,7240:724012-724012-11.
[22] Zhang Lin, Zhang Lei, Mou Xuanqin and Zhang David. FSIM: A Feature SimilarityIndex for Image Quality Assessment [J]. IEEE Transactions on Image Processing,2012,20(8):2378-2385.
[23] Rehman A, Wang Z. Reduced-reference image quality assessment by structural similarityestimation [J]. IEEE Transactions on Image Processing,2012,21(8):3378-3389.
[24] Wang Z, Simoncelli E P, Bovik A C. Multiscale structural similarity for image qualityassessment[C]. In: Signals, Systems and Computers,2003. Conference Record of theThirty-Seventh Asilomar Conference on. IEEE,2003,2:1398-1402.
[25] Wang Z, Shang X. Spatial pooling strategies for perceptual image quality assessment[C].In:Image Processing,2006IEEE International Conference on. IEEE,2006:2945-2948.
[26] Bovik E A. Handbook of Image and Video Processing [M]. Second Edition. ElsevierAcademic Press,2002
[27] Wang Z, Simoncelli E P. Reduced-reference image quality assessment using awavelet-domain natural image statistic model [J]. In Proc. SPIE Human Vision andElectronic Imaging X,2005,5666(1):149-159
[28] Gao X B, Lu W, Li X L, et al. Image quality assessment based on multiscale geometricanalysis [J]. IEEE Trans. Image Processing,2009,18(7):1608-1622
[29] Lu W, Gao X, Li X, et al. An image quality assessment metric based contourlet[C].In:Image Processing,2008. ICIP2008.15th IEEE International Conference on. IEEE,2008:1172-1175.
[30]路文,高新波,王体胜.一种基于WBCT的自然图像质量评价方法[J].电子学报,2008,36(2):303-308
[31] Zeng K, Wang Z. Polyview Fusion: A Strategy to Enhance Video-Denoising Algorithms[J]. Image Processing, IEEE Transactions on,2012,21(4):2324-2328.
[32] Yim C, Bovik A C. Quality assessment of deblocked images [J]. Image Processing, IEEETransactions on,2011,20(1):88-98.
[33] Kim S B, Wang Z, Hiremath B. A Bayesian approach for the alignment of high-resolutionNMR spectra [J]. Annals of Operations Research,2010,174(1):19-32.
[34] Ferzli Rony, Karam L J. A No-Reference Objective Image Sharpness Metric Based on theNotion of Just Noticeable Blur (JNB)[J]. IEEE Transactions on Image Processing,2009,18(4):717-728
[35] Pons A, Malo J, Artigas M J and Capilla. Image quality metric based onmultidimensional contrast perception models [J]. Displays,1999,20:93–110.
[36] Wee C Y, Paramesran R, Mukundan R, et al. Image quality assessment by discreteorthogonal moments [J]. Pattern Recognition,2010,43(12):4055-4068.
[37] He L, Gao X, Lu W, et al. Image quality assessment based on S-CIELAB model [J].Signal, Image and Video Processing,2011,5(3):283-290.
[38] Capodiferro L, Jacovitti G, Di Claudio E D. Two-Dimensional Approach toFull-Reference Image Quality Assessment Based on Positional Structural Information [J].Image Processing, IEEE Transactions on,2012,21(2):505-516.
[39] He L, Tao D, Li X, et al. Sparse representation for blind image quality assessment[C]. In:2012IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2012:1146-1153.
[40] Sheikh H R, Bovik A C. A visual information fidelity approach to video qualityassessment[C]. In:The First International Workshop on Video Processing and QualityMetrics for Consumer Electronics.2005:23-25.
[41] Narvekar N D, Karam L J. A no-reference image Blur metric based on the cumulativeprobability of Blur detection (CPBD)[J]. IEEE Transactions on Image Processing,2011,20(9):2678-2683
[42] Li C F, Yuan W, Bovik A C and Wu X. No-reference Blur index using Blur comparisons[J].Electronics Letters,2011,47(17):962-963.
[43] Wang Z, Sheikh H R, Bovik A C. No-reference perceptual quality assessment of JPEGcompressed images[C].In: International Conference on Image Processing.2,2002,1:I-477-I-480.
[44] Narwaria M, Lin W, McLoughlin I V, et al. Fourier transform-based scalable imagequality measure [J]. Image Processing, IEEE Transactions on,2012,21(8):3364-3377.
[45] Liu S, Bovik A C. Efficient DCT-domain blind measurement and reduction of blockingartifacts. IEEE Trans [J]. Circuits and System for Video Technology,2002,12(12):1139-1149
[46] Watson A B. DCTune: A technique for visual optimization of DCT quantizationmatrices for individual images[C]. In:Sid International Symposium Digest of TechnicalPapers. SOCIETY FOR INFORMATION DISPLAY,1993,24:946-946.
[47] Sheikh H R, Bovik A C, and Cormack L. No-reference quality assessmentusing naturalscene statistics: JPEG2000[J]. IEEE Trans. Image Processing,2005,14(11):1918-1927
[48] Wang Z and Bovik A C. Bitplane-by-bitplane shift (BbBShift)-A suggestion for JPEG2000region of interest coding [J].IEEE Signal Processing Letters,2002,9(5):60–162
[49] Marziliano P, Dufaux F, Winkler S and Ebrahimi T. Perceptual blur and ringing metrics:Application to JPEG2000[J]. Signal Processing: Image Communication,2004,19(2):163–172
[50] Li C, Bovik A C and Wu X. Blind image quality assessment using a general regressionneural network [J]. IEEE Trans. Neural Netw,2011,22(5):793–799
[51] Moorthy A K and Bovik A C. Blind image quality assessment: from scene statistics toperceptual quality [J]. IEEE Trans. Image Process,2011,20(12):3350–3364
[52] Saad M A, Bovik A C and Charrier C. Model-based blind image quality assessment: anatural scene statistics approach in the DCT domain [J]. IEEE Trans. Image Process,2012,21(8):3339–3352
[53] Moorthy A K and Bovik A C. A two-step framework for constructing blind image qualityindices [J]. IEEE Signal Process. Lett,201017(5):513–516
[54] Mittal A, Muralidhar G S and Bovik A C. Blind image quality assessment without humantraining using latent quality factors [J]. IEEE Signal Process. Lett,2012,19(2):75–78
[55] Mittal A, Muralidhar G S and Bovik A C. Making a ‘completely blind’ image qualityanalyzer [J]. IEEE Signal Process. Lett,2013,20(3):209–212
[56] Li Chaofeng, Ju Yiwen, Bovik A C, et al. No-training, no-reference image quality indexusing perceptual features [J]. Optical Engineering,2013
[57] Xue W, Zhang L, Mou X. Learning without Human Scores for Blind Image QualityAssessment [C]. In: IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2013.995-1001.
[58] Video Quality Experts Group [DB/OL]. http://www.vqeg.org/
[59] Sheikh H R, Sabir M F and Bovik A C. A statistical evaluation of recent full referenceimage quality assessment algorithms [J]. IEEE Trans. Image Processing,2006,15(11):3440–3451
[60] Serir A, Beghdadi A, Kerouh F. No-Reference Blur Image Quality Measure Based onMultiplicative Mutiresolution Decomposition [J]. Journal of Visual Communication andImage Representation,2013,24(7):911-925.
[61] Larson E C and Chandler D M. Categorical Image Quality (CSIQ) database2009
[DB/OL]. http://vision.okstate.edu/csiq
[62] Sheikh H R, Seshadrinathan K, et al.. Image and Video Quality Assessment Researce atLIVE2004[DB/OL]. http://live.ece.utexas.edu/research/quality
[63] Ponomarenko N, Lukin V, Zelensky A, et al.. TID2008—A database for evaluation offull-reference visual quality assessment metrics [J]. Adv. Modern Radioelectron,2009,10:30–45
[64] Ninassi A, Le Callet P, and Autrusseau F. Subjective Quality Assessment-IVC Database2005[DB/OL]. http://www2.irccyn.ecnantes.fr/ivcdb
[65] Horita Y, Shibata K, Kawayoke Y, and Sazzad Z M P. MICT Image Quality EvaluationDatabase2000[DB/OL]. http://mict.eng.u-toyama.ac.jp/mict/index2.html
[66] BID-Blurred Image Database[DB/OL]http://www.lps.ufrj.br/profs/eduardo/I-mageDatabase.htm
[67] Wang Z, Simoncelli E P. Reduced-reference image quality assessment using awavelet-domain natural image statistic model[C]. In:Proc. of SPIE Human Vision andElectronic Imaging.2005,5666:149-159.
[68] Gao X, Lu W, Tao D and Li X. Image quality assessment based on multiscale geometricanalysis [J]. IEEE Transactions on Image Processing,2009,18(7):1409–1423.
[69] Ma Lin, Lin Weisi, Deng Chenwei, et al. Image retargeting quality assessment： a studyof subjective scores and objective metrics[J]. IEEE Journal of Selected Topics In SignalProcessing,2012,6(6):626–639
[70]叶盛楠,苏开娜等.基于结构信息提取的图像质量评价[J].电子学报，2008,36(5):856-861
[71]杨春玲,高文瑞.基于结构相似的小波域图像质量评价方法的研究[J].电子学报，2009,37(4):845-849
[72]路文,高新波等.一种基于WBCT的自然图像质量评价方法[J].电子学报，2008,36(2):303-308
[73] Hassen R, Wang Z, Salama M. No-reference image sharpness assessment based on localphase coherence measurement[C]. In: IEEE International Conference on Acoustics Speechand Signal Processing (ICASSP),2010:2434-2437.
[74] Chen M J and Bovik A C. No-reference image blur assessment using multiscale gradient[J]. EURASIP Journal on Image and Video Processing,2011,3:1-11
[75] Ciancio A, Da Costa A L N T, et al. No-reference blur assessment of digital picturesbased on multi-feature classifiers [J]. IEEE Trans. Image Processing,2011,20(1):64-75
[76] Saad M A, Bovik A C and Charrier C. A DCT statistics-based blind image quality index[J]. IEEE Signal Process. Lett.,2010,17(6):583–586
[77] Wang S, Rehman A, Wang Z, et al. Ssim-motivated rate-distortion optimization for videocoding[J]. IEEE Transactions on Circuits and Systems for Video Technology,2012,22(4):516-529.
[78] Watson A B, Yang G Y, Solomon J A and Villasenor J. Visibility of wavelet quantizationnoise [J]. IEEE Trans. Image Processing,1997,6(8):1164–1175
[79] Chandler D M, Hemami S S. Additivity models for suprathreshold distortion in quantizedwavelet-coded images[C]. In: International Society for Optics and Photonics ElectronicImaging,2002:105-118.
[80] Selesnick I W, Baraniuk R G, Kingsbury N C. The dual-tree complex wavelettransform[J]. IEEE Signal Processing Magazine,2005,22(6):123-151.
[81] Portilla J, Wainwright V S M J and Simoncelli E P. Image denoising using scale mixturesof gaussians in the wavelet domain [J]. IEEE Trans. Signal Processing,2003,12(11):1331–1338
[82] Brunet D, Vrscay E R, Wang Z. On the mathematical properties of the structuralsimilarity index[J]. Image Processing, IEEE Transactions on,2012,21(4):1488-1499.
[83] Brunet D, Vass J, Vrscay E R, et al. Geodesics of the Structural Similarity index[J].Applied Mathematics Letters,2012,25(11):1921-1925.
[84] Wang Z, Li Q. Information content weighting for perceptual image quality assessment[J].Image Processing, IEEE Transactions on,2011,20(5):1185-1198.
[85] Rostami M, Michailovich O, Wang Z. Image deblurring using derivative compressedsensing for optical imaging application [J]. IEEE Transactions on Image Processing,2012,21(7):3139-3149.
[86] Nikvand N, Wang Z. Image distortion analysis based on normalized perceptualinformation distance [J]. Signal, Image and Video Processing, special issue on HumanVision and Information Theory,2013,7(3):403-410.
[87] Morrone M C and Owens R A. Feature detection from local energy. Pattern RecognitionLetters,1987,6(5)：303–313
[88] Mittal A, Moorthy A K and Bovik A C. No-Reference Image Quality Assessment in theSpatial Domain [J]. IEEE Trans. Image Process.,2012,21(12):4695–4708.
[89] Kovesi P. Image features from phase congruency. Computer Vision Research,1999,1(3):1-26
[90] Peter Kovesi homepage[EB/OL]. http://www.csse.uwa.edu.au/~pk/Research/MatlabFns/index.html
[91] Wang Z and Simoncelli E P. Local phase coherence and the perception of blur [J]. In Adv.Neural Information Processing Systems (NIPS03),2004,16:(5)
[92] Oppenheim A V, Lim J S. The importance of phase in signals [J]. Proceedings of theIEEE,1981,69(5):529-541.
[93] Morrone M C and Burr D C. Feature detection in human vision: A phase dependentenergy model [J]. Proc. R. Soc. Lond. Biological Sciences,1988,23(12):221–245.
[94] Kovesi P. Phase congruency: A low-level image invariant [J]. Psych.Research,2000,64:136–148
[95] Wang Z. Applications of objective image quality assessment methods [J]. IEEE SignalProcessing Magazine,2011,28(6):137-142.
[96] Hassen R, Wang Z, Salama M. Image Sharpness Assessment Based on Local PhaseCoherence[J] IEEE Transactions on Image Processing,2013,22(7):2789-2810.
[97] Haralick R M, Shanmugam K and Dinstein I. Textural Features for Image Classification[J]. IEEE Transactions on Systems, Man, and Cybernetics1973, SMC-3(6):610–621
[98] Hralick R M. Statistical and Structural Approaches toTexture[J]. Proceedings of IEEE,1979,67(5):786-805
[99] Vapnik V. The Nature of Statistical Learning Theory [M]. Berlin, Germany: SpringerVerlag,2000
[100] Narwaria M and Lin W. Objective image quality assessment based on support vectorregression[J]. IEEE Trans. Neural Netw.,2010,21(3):515–519
[101] Yeganeh H and Wang Z. Objective quality assessment of tone mapped images [J]. IEEETransactions on Image Processing,2013,22(2):657-667.
[102] Zhao T, Kwong S, Wang H, et al. Multiview Coding Mode Decision With HybridOptimal Stopping Model [J]. IEEE Transactions on Image Processing,2013,22(4):1598-1609.
[103] Wang S, Rehman A, Wang Z, et al. Perceptual Video Coding Based on SSIM-InspiredDivisive Normalization [J]. IEEE Transactions on Image Processing,2013,22(4):1418-1429.
[104] Varghese G, Wang Z. Video denoising based on a spatiotemporal Gaussian scalemixture model[J]. IEEE Transactions on Circuits and Systems for Video Technology,2010,20(7):1032-1040.
[105] Lyu S, Simoncelli E P. Statistical modeling of images with fields of Gaussian scalemixtures[C]. In: Advances in Neural Information Processing Systems.2006:945-952.
[106] Ye P, Kumar J, Kang L, et al. Unsupervised feature learning framework for no-referenceimage quality assessment [C]. In:2012IEEE Conference on Computer Vision and PatternRecognition (CVPR),2012:1098-1105.
[107] Chih-Jen Lin’s Home Page [EB/OL]. http://www.cise.ntu.edu.tw/~cjlin/libsvm
[108] Bovik A C. The Essential Guide to Image Processing [M]. New York：Academic Press,2009,7:421-426.
[109] Specht D F. A general regression neural network [J]. IEEE Trans. Neural Netw.,1991,2(6):568-576
[110] Chartier S, Boukadoum M and Amiri M. BAM learning of nonlinearly separable tasksby using an asymmetrical output function and reinforcement learning [J]. IEEE Trans.Neural Netw.,2009,20(8):1281–1292.
[111] Tang H, Joshi N, Kapoor A. Learning a blind measure of perceptual image quality [C].In:2011IEEE Conference on Computer Vision and Pattern Recognition (CVPR),2011:305-312.
[112] Chetouani A, Beghdadi A, Chen S and Mostafaoui G. A novel free reference imagequality metric using neural network approach [J]. In Proc.Int. Workshop Video Process.Qual. Metrics Cons. Electron., Scottsdale,AZ,2010(1):1–4.
[113] Li Q, Meng Q, Cai J, Yoshino H and Mochida A. Predicting hourly cooling load in thebuilding: A comparison of support vector machine and different artificial neural networks[J]. Ener. Conv. Manage.,2009,50(1):90–96
[114] Aleksandr S, Alexander G, and Ahmet M E. An SVD-Based Grayscale Image QualityMeasure for Local and Global Assessment [J]. IEEE Trans. Image Process.,2006,15(2):422-429
[115] Narwaria M, Lin W. Objective image quality assessment with Singular ValueDecomposition[C]. In: Proc.5th Int. Workshop Video Processing and Quality Metrics forConsumer Electronics.2010.
[116]楼斌，沈海斌，赵武锋等．基于自然图像统计的无参考图像质量评价[J]．浙江大学学报：工学版，2010,44(2):248-252.
[117]金波，李朝锋，吴小俊．结合NSS和小波变换的无参考图像质量评价[J].中国图象图形学报，2012,17(1):33-39
[118] Cheng C, Wang H. Quality assessment for color images with tucker decomposition[C].In:201219th IEEE International Conference on Image Processing (ICIP),2012:1489-1492.
[119] Abdi H,Williams L J. Principal component analysis.Wiley Interdisciplinary Reviews:Computational Statistics,2010,2:433–459
[120] Andrews H C, Patterson C L. Singular Value Decomposition Image Coding [J]. IEEETrans. on Communications,1976(4):425-432.
[121] Pyatykh S, Hesser J, Zheng L. Image noise level estimation by principal componentanalysis [J]. IEEE Trans. Image Process.,2012,22(12):5226-5237.