基于盲解卷积的图像盲复原技术研究
|
摘要
航空遥感成像、医学成像等成像领域由于成像条件的复杂性会使得图像的分辨率和对比度降低。因此，图像复原一直以来是计算机视觉、图像处理的研究热点。大多数图像复原算法如维纳滤波复原等都是建立在对图像的点扩散函数已知的前提下，而现实中的成像条件如相机的相对运动、大气扰动等都是未知的，它们所造成的点扩散函数是无法精确得知的，因此对盲解卷积及盲复原技术理论及算法的研究，有着现实和理论的需求。
本论文主要针对先验信息未知或者部分已知的模糊图像，从图像盲复原和图像盲评价两方面着手，详细地研究了其理论基础和关键技术。对于盲复原，分别论述了模糊类型已知和模糊类型未知两种情况，对其中的参数估计及正则化技术等关键问题进行了深入研究；对于盲评价，分别研究了噪声、模糊度、块效应的无参考评价方法，以及由这三个失真因素所构成的综合评价方法。主要创新工作和研究成果如下：
1.在模糊类型已知方面，针对单一模糊类型，先建立点扩散函数参数模型并通过关于图像本身或成像系统的先验知识估计其参数，然后使用估计后的点扩散函数模型实现模糊图像的复原。在此理论的基础上，本文重点研究了离焦模糊的特性，提出了一种基于Hough变换估计离焦模糊半径的新方法。该方法具有估计精度高和稳定性高的优点，能有效地用于图像复原过程中的事先参数估计中。
2.在模糊类型未知方面，将点扩散函数估计和图像复原同时进行，提出了一种自适应阈值的超变分正则化盲复原算法。该算法一方面提出了自适应阈值的超变分正则项，并且根据图像噪声的标准差选取阈值，使得图像复原过程中，边缘区得以保持，平滑区得以去噪；另一方面，使用辅助变量及半二次规整理论，克服了传统求解中过多地使用近似的缺点，求出了该模型最优化式的精确解。实验结果表明，该算法使得恢复后的图像细节增加且块效应明显减少。
3.分析了现有的无参考图像质量评价方法，针对这些方法无法与主观评价一致的缺点，提出了一种新的无参考图像质量评价方法：基于噪声、模糊度和块效应的无参考图像质量评价方法。该方法考虑了图像失真的3种主要类型：噪声、模糊度和块效应。首先利用修正的小波中值估计法得到图像噪声的标准差；接着利用图像边缘像素点的方法来反映图像的模糊度；然后利用图像像素块的特征来表征图像的块效应；最后综合上述3种失真类型，构造了图像质量评价模型。我们结合LIVE IQADatabase所提供的主观评价值（DMOS），使得该算法不仅在一定程度上契合PSNR，并且也与主观评价（DMOS）值保持一致。
4.从算法的稳定性、算法的高效性、算法的应用、图像的视觉效果等主客观评价标准，以及PSNR、图像熵等图像质量评价标准，对实验结果进行了评价和分析。给出了实验结果图像和实验数据值，并与同类型的算法进行了复原效果和评价效果的对比分析。
In remote sensing imaging and medical imaging, image resolution and contrastwill be decreased because of the complexity of imaging conditions. So imagerestoration has being a research focus in computer vision and image processing.Existing image restoration algorithms, such as Wiener filtering, built on the knownpoint spread function. However，point spread function were often unknown in practicebecause imaging conditions were unknown, like the relative motion between cameraand objects and atmospheric disturbance. Therefore, the research on blinddeconvolution and blind restoration is very necessary in theory and practice.
This dissertation was mainly to the blurred images whose prior information wasunknown or partially known. We detailed researched on the basic theories and keytechnologies in two spects of image blind restoration and image blind assessment. Forimage blind restoration, we discussed two spects of blur type being known andunknown, deeply researching on key problems about parameter estimation andregularization technique. For image blind assessment, we respectively researched onno-reference image assessment algorithms of noise, blur degree and blocking effectsand comprehensive evaluation algorithm that was comprised of three distortionfactors. The main innovations and research results are as following:
1. For the known blur type, this dissertation only studied the single blur type. Wefirstly established the parameter model of point spread function, estimated its parameters by the prior knowledge about degraded image and imaging system andthen restored the blurred image by using the estimated point spread function model.Based on this theory, after mainly studying the characteristics of defocused blurredimage, a novel scheme, which was based on the Hough transform, was proposed forestimating the radius of point spread function of defocused blurred images. Thisscheme was highly accurate and highly stable. It’s very useful for parametersestimation in the processing of image restoration.
2. For the unknown blur type, the estimation of point spread function and imagerestoration were carried on simultaneously. A super total variation image blinddeblurring method with self adaptive threshold was proposed to restore the imagesdegraded by unknown point spread function. On the one hand, based on the analysisof the total variation model, the super total variation with self adaptive threshold wasproposed and the threshold was deduced by the estimation noise of image. In the case,the edge areas of image were preserved and the smooth areas were denoised in theprocessing of image restoration. On the other hand, the exact solution of optimizationmodel was obtained by using semi-quadratic regularization and auxiliary variables,which overcome shortcomings of using too many approximations in many traditionalsolutions. The experimental results demonstrated that the restoration image has moredetails and fewer blocking effects.
3. This dissertation analyzed existing no-reference image quality assessmentalgorithms. Since these algorithms were not consistent with subjective assessment, anovel no-reference image quality assessment method which was based on noise，blurdegree and blocking effects was proposed. It introduced three types of imagedistortion, including noise, blur degree and blocking effects. Firstly, the standarddeviation of image noise was estimated by modified wavelet medium estimation.Secondly, the blur degree of image was obtained by using edge pixel points. Thirdly,blocking effects was represented by using characteristics of image pixel blocks.Finally, the assessment model was established by combining foregoing threedistortion types. Combining the Differential Mean Opinion Scores (DMOS) provided in the LIVE IQA DataBase, The evaluation values of this algorithm not only agreedwith PSNR in objective assessment, but also were consistent with the DMOS insubjective assessment.
4. Combining the objective and subjective image quality assessment, thisdissertation analyzed and evaluated the experimental results from the stability, theefficiency and the application of algorithm, visual effects of images, as well as imageevaluation indices, like PSNR, image entropy. The experimental images and data werepresented. Also, this dissertation compared our algorithms with the same typealgorithms in effect of image restoration and image quality assessment.
本论文主要针对先验信息未知或者部分已知的模糊图像，从图像盲复原和图像盲评价两方面着手，详细地研究了其理论基础和关键技术。对于盲复原，分别论述了模糊类型已知和模糊类型未知两种情况，对其中的参数估计及正则化技术等关键问题进行了深入研究；对于盲评价，分别研究了噪声、模糊度、块效应的无参考评价方法，以及由这三个失真因素所构成的综合评价方法。主要创新工作和研究成果如下：
1.在模糊类型已知方面，针对单一模糊类型，先建立点扩散函数参数模型并通过关于图像本身或成像系统的先验知识估计其参数，然后使用估计后的点扩散函数模型实现模糊图像的复原。在此理论的基础上，本文重点研究了离焦模糊的特性，提出了一种基于Hough变换估计离焦模糊半径的新方法。该方法具有估计精度高和稳定性高的优点，能有效地用于图像复原过程中的事先参数估计中。
2.在模糊类型未知方面，将点扩散函数估计和图像复原同时进行，提出了一种自适应阈值的超变分正则化盲复原算法。该算法一方面提出了自适应阈值的超变分正则项，并且根据图像噪声的标准差选取阈值，使得图像复原过程中，边缘区得以保持，平滑区得以去噪；另一方面，使用辅助变量及半二次规整理论，克服了传统求解中过多地使用近似的缺点，求出了该模型最优化式的精确解。实验结果表明，该算法使得恢复后的图像细节增加且块效应明显减少。
3.分析了现有的无参考图像质量评价方法，针对这些方法无法与主观评价一致的缺点，提出了一种新的无参考图像质量评价方法：基于噪声、模糊度和块效应的无参考图像质量评价方法。该方法考虑了图像失真的3种主要类型：噪声、模糊度和块效应。首先利用修正的小波中值估计法得到图像噪声的标准差；接着利用图像边缘像素点的方法来反映图像的模糊度；然后利用图像像素块的特征来表征图像的块效应；最后综合上述3种失真类型，构造了图像质量评价模型。我们结合LIVE IQADatabase所提供的主观评价值（DMOS），使得该算法不仅在一定程度上契合PSNR，并且也与主观评价（DMOS）值保持一致。
4.从算法的稳定性、算法的高效性、算法的应用、图像的视觉效果等主客观评价标准，以及PSNR、图像熵等图像质量评价标准，对实验结果进行了评价和分析。给出了实验结果图像和实验数据值，并与同类型的算法进行了复原效果和评价效果的对比分析。
In remote sensing imaging and medical imaging, image resolution and contrastwill be decreased because of the complexity of imaging conditions. So imagerestoration has being a research focus in computer vision and image processing.Existing image restoration algorithms, such as Wiener filtering, built on the knownpoint spread function. However，point spread function were often unknown in practicebecause imaging conditions were unknown, like the relative motion between cameraand objects and atmospheric disturbance. Therefore, the research on blinddeconvolution and blind restoration is very necessary in theory and practice.
This dissertation was mainly to the blurred images whose prior information wasunknown or partially known. We detailed researched on the basic theories and keytechnologies in two spects of image blind restoration and image blind assessment. Forimage blind restoration, we discussed two spects of blur type being known andunknown, deeply researching on key problems about parameter estimation andregularization technique. For image blind assessment, we respectively researched onno-reference image assessment algorithms of noise, blur degree and blocking effectsand comprehensive evaluation algorithm that was comprised of three distortionfactors. The main innovations and research results are as following:
1. For the known blur type, this dissertation only studied the single blur type. Wefirstly established the parameter model of point spread function, estimated its parameters by the prior knowledge about degraded image and imaging system andthen restored the blurred image by using the estimated point spread function model.Based on this theory, after mainly studying the characteristics of defocused blurredimage, a novel scheme, which was based on the Hough transform, was proposed forestimating the radius of point spread function of defocused blurred images. Thisscheme was highly accurate and highly stable. It’s very useful for parametersestimation in the processing of image restoration.
2. For the unknown blur type, the estimation of point spread function and imagerestoration were carried on simultaneously. A super total variation image blinddeblurring method with self adaptive threshold was proposed to restore the imagesdegraded by unknown point spread function. On the one hand, based on the analysisof the total variation model, the super total variation with self adaptive threshold wasproposed and the threshold was deduced by the estimation noise of image. In the case,the edge areas of image were preserved and the smooth areas were denoised in theprocessing of image restoration. On the other hand, the exact solution of optimizationmodel was obtained by using semi-quadratic regularization and auxiliary variables,which overcome shortcomings of using too many approximations in many traditionalsolutions. The experimental results demonstrated that the restoration image has moredetails and fewer blocking effects.
3. This dissertation analyzed existing no-reference image quality assessmentalgorithms. Since these algorithms were not consistent with subjective assessment, anovel no-reference image quality assessment method which was based on noise，blurdegree and blocking effects was proposed. It introduced three types of imagedistortion, including noise, blur degree and blocking effects. Firstly, the standarddeviation of image noise was estimated by modified wavelet medium estimation.Secondly, the blur degree of image was obtained by using edge pixel points. Thirdly,blocking effects was represented by using characteristics of image pixel blocks.Finally, the assessment model was established by combining foregoing threedistortion types. Combining the Differential Mean Opinion Scores (DMOS) provided in the LIVE IQA DataBase, The evaluation values of this algorithm not only agreedwith PSNR in objective assessment, but also were consistent with the DMOS insubjective assessment.
4. Combining the objective and subjective image quality assessment, thisdissertation analyzed and evaluated the experimental results from the stability, theefficiency and the application of algorithm, visual effects of images, as well as imageevaluation indices, like PSNR, image entropy. The experimental images and data werepresented. Also, this dissertation compared our algorithms with the same typealgorithms in effect of image restoration and image quality assessment.
引文
[1] Nagel M R, Introduction to “Evaluation of Motion-Degraded Images” Proc. ofNASA Electronics Research Center Seminar, Cambridge, MA,1968Dec:3-5.
[2] Andrews.H.C, Hunt.B.R. Digital Image Restoration[M]. Englewood Cliffs NewJersey: Prentice-Hall,1977.
[3]邹谋炎.反卷积和信号复原[M].北京,国防工业出版社，2001.4-7
[4] R C Gonzalez, R E Woods.数字图像处理（第二版）[M].阮秋琦等译.北京：电子工业出版社，2007.
[5] Kundur D, Hatzinakos D. Blind image deconvolution revisited[J]. IEEETransactions on signal processing.1996,13(6):61-63.
[6] Q Shan, J Jia, A Agarwala. High-quality motion deblurring from a singleimage[J]. ACM Transaction on Graphics,2008,27(3)：1-10.
[7]邓泽峰，熊有伦.基于频域方法的运动模糊方向识别[J].光电工程.2007,34(10):98-101.
[8] J F Cai, H Ji, C Liu, et al. Blind motion deblurring from a single image usingsparse approximation[C]. IEEE Conference on Computer Vision and PatternRecognition, Miami Beach, FL:IEEE Press,2009:104-111.
[9]李铁成,陶小平,冯华君等.基于倾斜刃边法的调制传递函数计算及图像复原[J].光学学报,2010,30(10):2891-2897.
[10] P J Bex, W Makous. Spatical frequcncy, phase, and the contrast of naturalimages[J]. Journal of the optical society of America,2002,19(6):1096-1106.
[11] J.Da Rugna, H.Konik. Automatic blur detection for metadata extraction incontent-based retrival context[C]. In SPIE, volume5304, pages285-294,2003.
[12] Renting Liu, Zhaorong Li, Jiaya Jia. Image Partial Blur Detection andClassification[C]. CVPR08(1-8). IEEE DOI Link0806.
[13] R.L.Lagendijk, J.Biemod, D.E.Boekee. Identification and restoration of noisyblurred images using the Expectation-Maximization algorithm[J]. IEEETransactions on signal processing.1990,38(7):1180-1190.
[14] R.G. Lane, R.H.T.Bates.Automatic multidimensional deconvolution [J].Journal of the Optical Society.1987,4(1):180-188.
[15] Ayers G.R, Dainty J.C. Iterative blind deconvolution method and itsapplications [J]. Optics Letters,1998,13(7):547-549.
[16] McCallum B C. Bind deconvolution of speckle images[J]. J.Opt.Soc.Am., A.1987,4:180-188.
[17] Rudin L, Osher S, Fatemi E, Nonlinear total variation based noise removalalgorithms[J]. Physica D.1992,60:259-268.
[18] T.F.Chan, C.K.Wong, Total variation blind deconvolution[J].19984(3):370-375.
[19] Esedoglu S, Osher S. Decomposition of images by the anisotropicRudin-Osher-Fatemi model [J]. Communication on Pure and Applied Mathematics.2004,57:1609-1626.
[20] Haiyong Liao, Michael K. Ng. Blind Deconvolution Using GeneralizedCross-Validation Approach to Regularization Parameter Estimation[J]. IEEETransactions on Image Processing,2011,20(3):670-680.
[21] Giannakis Georgios B, Heath Robert W. Blind identification of multi-channelFIR blurs and perfect image restoration[J]. IEEE Transactions on Image Processing,2000,9(11):1877-1896.
[22] J F Cai, H Ji, C Liu, et al. Blind motion deblurring using multiple images[J].Journal of Computational Physics,2009,228(14):5057-5071.
[23] M.Van Noort, L.R. Vaner Voort, M.G.Lofdahl. Solar image restoration by useof multi-frame blind deconvolution with multiple objects and phase diversity[J].Sol. Phys.228,191-215,2005.
[24] A bd-krim Seghouane. A kullback-Leibler divergence approach to blind imagerestoration[J]. IEEE Transactions on Image processing.20(7),2011.
[25] J.H.Money, S.H.Kang. Total variation minimizing blind deconvolution withshock filter reference. Image vis. Comput,26(2):302-314,2008.
[26] L.Bar, N.Sochen, N.Kiryati. Variational pairing of image segmentation andblind restoration[J]. in Proc.8th Eur. Conf. Computer Vision, Prague, Czech Public,May,2004, vol. Part II, LNCS3022:166-177.
[27] M.S.C. Almeida, L.B.Almeida. Blind deblurring of natural images. Presentedat the ICASSP, Las Vegas, NV, Apr,2008.
[28] H. R. Wu, and M. Yuen, A generalized block-edge impairment metric for videocoding[J]. IEEE Signal Proc.Letters, vol.4, no.11, pp.317-320, Nov.1997.
[29] D. Wei and A. C. Bovik, A new metric for blocking artifacts in blocktransform-coded images and video[J]. Technical Report, The University of Texas,Oct.1998.
[30] Hamid Rahim Sheikh, Alan Conrad Bovik. No-Reference Quality AssessmentUsing Natural Scene Statistics: JPEG2000[J]. IEEE Transactions on imageprocessing, vol.14, no.11, nov2005.
[31] Alan C,Bovki, Shizhong Liu. DCT-domain blind measurement of blockingartifacts in DCT-coded images[C]. IEEE International Conference onAcoustics,Speech and Signal Processing. Slat Lake City,2001,1725-1728.
[32]张航，罗大庸.图像盲复原算法研究现状及其展望[J].中国图象图形学报,2004.9(10):1145-1152.
[33] M.Bronstein, A.Bronstein, M.Zibulevsky and Y.Zeevi. Blind deconvolution ofimages using optimal sparse representations[J]. IEEE Transaction on imageprocessing, vol.14, no.6, pp.726-736, Jun.2005.
[34] Deng Zhipo, Xia Youshen. A neural network algorithm for fast blind imagerestoration using a novel2D-ARMA parameter estimation[J]. Internationalconference on Audion Language and Image Processing,2010.
[35] T.A.Cheema, I.M.Qureshi, A.Hussain. Blind image deconvolution usingspace-variant neural network approach[J]. Electronics Letters.41(6):308-309,2005.
[36] Lun Daniel P K, Hsung T C, Shen T W. Orthogonal discrete periodic Radontransform. Part I: Theory and realization[J]. Signal Processing,2003,83(5):941-955.
[37] Lun Daniel P K, Hsung T C, Shen T W. Orthogonal discrete periodic Radontransform. Part II: applications[J]. Signal Processing,2003,83(5):957-971.
[38] Y. Wang, W.Yin, Y.Zhang, A fast algorithm for image deblurring with totalvariation regularization[J]. CAAM Tech. Rep. TR07-10(2007), Rice Univ., Houston,TX,2007.
[39]李仕.航空异速像移模糊实时恢复算法研究与GPU平台实现[D].［博士学位论文］．长春：中国科学院长春光学精密机械与物理研究所,2010.
[40]沈峘,李舜酩,毛建国,辛江慧.数字图像复原技术综述[J].中国图象图形学报.2009.
[41]颜昌翔，王家琪.航空相机像移补偿计算的坐标变换方法[J].光学精密工程，2000，8(3):203-207.
[42] Reeves S I, Mersereau R M, Blur identification by the method of generalizedcross-validation[J]. IEEE Transactions on Image Processing,1992,1(3):301-311.
[43]史光辉.像移对卫星摄影成像质量的影响[J].光学精密工程，1997,5(4):31-34
[44]王晓红,赵荣椿.匀速直线运动模糊的PSF之估计[J].计算机应用.2001,21(9):40-41.
[45]王晓红,赵荣椿.任意方向运动模糊的消除[J].中国图象图形学报.2000,5(6):525-529.
[46]苏光大.图像并行处理技术[M].清华大学出版社.2002,64-65.
[47] Stokseth P.A. Properties of a defocused optical system [C], J. Opt. Soc. Am.1969,59:1314-1321.
[48] J.W.Hardy. Adaptive optics for astronomical telescopes[M]. Oxford UniversityPress, New York,1998.
[49] M.C.Roggemann and B.Welsh. Imagine Through Turblence [M]. CTC Press,Boca Raton, FL,1996.
[50] C.Vogel. Computational Methods for Inverse Problems[J]. AppliedMathematics. SIAM,2002.
[51]杨枝灵，王开. Visual C++数字图形获取处理及实践应用[M].北京：人民邮电出版社，2003.
[52] Sondhi M M, Image restoration: The removal of spatially invariant degrada-tions[J], Proceedings of the IEEE,1972,60(7):842-853.
[53]王旭辉,郭光亚.匀速运动模糊图像的快速恢复[J].微计算机信息.2000,16(5):5-7.
[54] R Hunt B. A Matrix Theory Proof of the Discrete Convolution Theorem[C].IEEE Trans.1971, AU-19(4):285-288.
[55] Jain A K, Fundamentals of Digital Image Processing[M]. Prentice-Hall,Englewood Cliffs, N J.,1989.
[56]贾平,张葆,孙辉.航空成像像移模糊恢复技术[J].光学精密工程,2006,14(4):697-703.
[57]张宏林. Visual C++数字图像处理典型算法及实现[M].人民邮电出版社,2008.
[58]刘微，运动模糊图像恢复算法的研究与实现[D],［博士学位论文］．长春：中国科学院长春光学精密机械与物理研究所,2006.
[59]王新成.高级图像处理技术[M].中国科学技术出版社,2001.
[60] Rafael Molina, Aggelos K. Katsaggelos, Javier Mateos. Bayesian andregularization methods for hyperparameter estimation in image restoration[J]. IEEETransactions on Image Processing.8(2):231-246.
[61] Kulmer S. Information Theory and statistics[M]. New York: Dover.
[62] Portilla J, Simoncelli E. Image restoration using Gaussian scale mixtures in thewavelet domain[A]. Proceedings of the IEEE Conference on Image Processing[C].IEEE Signal Processing Society,2004,2:965-968.
[63] Kundur D. Blind deconvolution of still images using recursive inversefiltering[J]. M.A.Sc. Thesis, University of Toronto, Department of electrical andComputer Engineering.1995,1-19.
[64] Kundur D, Hatzinakos D. A Novel blind deconvolution scheme for imagerestoration using recursive filtering[J]. IEEE Transactions on Signal Proeessing，1998，46(2):375-390.
[65] Kundur D, Hatzinakos D. On the use of lyapunov criteria to analyze theconvergence of blind deconvolution algorithms[J]. IEEE Transactions on SignalProeessing，1998，46(11):2918-2925.
[66]来沛剑，基于全变分的含噪图像盲复原算法研究[D],[硕士学位论文].昆明：昆明理工大学,2010.
[67] Rafael Molina. Javoer Mateos, Aggelos K. Katsaggelos. Blind deconvolutionusing a variational approach to parameter, image and blur estimation[J]. IEEE Trans.Image processing.15(12):3715-3727,2006.
[68] A.K.Katsaggelos, K.T.Lay. Maximum likelihood blur identification and imagerestoration using the EM algorithm[J]. IEEE Trans. signal processing.39(3):729-733,2004.
[69] A.C.Likas, N.P.Galatsanos. A variational approach for Bayesian blind imagedeconvolution[J]. IEEE Trans. signal processing.52(8):2222-2233,2004.
[70] Giannis Chantas, Nikolaos, Aristidis Likas, Michael Saunders. Variationalbayesian image restoration based on a product of t-distributions image prior[J].IEEE Trans. image processing,200817(10):1795-1805.
[71]魏政刚，袁杰辉，蔡元龙.图象质量评价方法的历史、现状和未来[J].中国图象图形学报，1998，3(5):386-389.
[72] Eskicioglu A M, Fisher P S, Image Quality Measures and Their Performance[J].IEEE Transactions on Communications.1995,43.
[73] Wang Z, Cormack L, et al. Laboratory for Image&Video Engineering[EB/OL]. http://live.ece.utexas.edu/research/quality,2002.
[74]林海祥，张炘.无参考图像质量评价综述[J].电脑知识与技术，2009,5(28):8043-8046.
[75] Wang Z, Bovik A C, A Universal Image Quality Index[J], IEEE SIGNALPROCESSING LETTERS.2002,9.
[76] Hosaka K，A New Picture Quality Evaluation Method[C], Proceedings of theInternational Picture Coding Symposium, April1986.
[77] Olsson S, Stropplan A M, Baina J,Objective methods for assessment of videoquality: state of the art [J]. IEEE Transactions on Broadcasting,1997,43(4):487-495
[78] Z.Wang, A.C.Bovik, H.R.Sheikh, et al. Image quality assessment: from errorvisibility to structural similarity. IEEE Transactions on Image Processing.2004,13(4):600-612.
[79]肖文.空域无参考的图像质量评价[D].[硕士学位论文].南昌：江西财经大学，2006.
[80]曹茂永，孙农亮，郁道银.基于灰度梯度的数字图像评价函数[J].光电工程，2003，30(6):69-72.
[81]胡良梅,高隽,何柯峰.图像融合质量评价方法的研究[J].电子学报.2004,32(12):218-221.
[82]丁艺芳.基于小波变换和视觉系统的图像质量综合评价新方法[D].[博士学位论文].上海：上海大学.2001.
[83]徐鲁安,章琦等.一种新的图像质量评价方法[J].计算机工程与设计.2004,25(3):418-420.
[84]刘艳春.利用传播波方程作运动模糊图像质量恢复的客观评价[D].[硕士学位论文].广州：暨南大学.2002.
[85]崔清亮.图像质量评价在虹膜识别中的应用研究[D].[硕士学位论文].长春：吉林大学.2004.
[86]魏政刚，袁杰辉,蔡元龙.一种基于视觉感知的图像质量评价方法[J].电子学报，1999，27(4):79-82.
[87] Osberger W, Maeder A J. A technique for image quality assessment based on ahuman visual systerm model. EUSIPCO-98.
[88]吴振宇，杜少军，姚洪利.采用聚焦度量的散焦模糊图像参数鉴别及复原[J].红外与激光工程，2011，40(4):772-776.
[89] Aihong Liu, Guosheng Hu, Haiyan Wu. A New Method for Blur Identificationof Out-of-Focus Images[J]. Education Technology and Computer Science,2009,298-303.
[90] Cannon.TM. Blind deconvolution of spatially invariant blurs with phase[C].IEEE Trans. on Acoustics Speech and signal processing,1976,24(1):58-63.
[91]穆丽娟.数字图像自动聚焦与离焦模糊复原技术研究[D].[硕士学位论文].天津：天津师范大学.2009.
[92]赵琳,金伟其,陈翼男等.基于微分图像自相关的离焦模糊图像盲复原[J].光学学报,2008,28,(9):1703-1709.
[93]陈前荣等.利用拉氏算子鉴别散焦模糊图像点扩散函数[J].计算机工程与科学.2005,27(9):40-43
[94]赵书安等. Hough变换在几何特征检测中的应用[J].南京师范大学学报（工程技术版）.2006,6(4):66-70.
[95]朱芳芳等.一种改进的Hough变换直线检测算法[J].计算机技术与发展.2009,19(5):19-22.
[96]刘宏申，程健.广义Hough变换算法的分析改进[J].中国科学技术大学学报.2009,39(11):1202-1206.
[97]杨全银，魏雅杰，王欣.基于Hough变换的线段检测算法[J].红外与激光工程，2009,38(3):525-529.
[98]秦开怀，王海颖，郑辑涛.一种基于Hough变换的圆和矩形的快速检测方法[J].中国图象图形学报.2010,15(1):109-115.
[99]杨全银.基于Hough变换的图像形状特征检测[D].[硕士学位论文].济南：山东大学.2009.
[100]王彬生，黄乡生. Hough变换及其在几何特征检测中的应用[J].计算机与现代化.2008,4:20-22.
[101] Tikhonov A N, Arsenin V Y. Solutions of ill-posed problem. New York:Wiley,1977:87-94
[102] Hadamard J. Lectures on the Cauchy problems in linear partial differentialequations. Yale University Press, New Haven.1923.
[103] Kirsch A. An Introduction to the Mathematical Theory of Inverse Problems.NewYork: Springer-Verlag,1996:5-11.
[104] R. L. Lagendijk, J. Biemond, D. E. Boekee. Regularized iterative imagerestoration with ringing reduction[J]. IEEE Transactions on Acoustics, Speech, andSignal Processing,1988,36(12):1874-1887.
[105]郭永彩,王婀娜,高潮.空间自适应和正则化技术的盲图像复原[J].光学精密工程，2008,16(11):2263-2267.
[106] Song H H, Zhang L, Wang P K, et al. An adaptive L1-L2hybrid error modelto superresolution[C]. Image Processing, IEEE International Conference,2010.2812-2824.
[107] S.Derin Babacan, Rafael Molina, Aggelos K.Katsaggelos. Variationalbayesian blind deconvolution using a total variation prior[J]. IEEE Transactions onimage processing,2009,18(1):12-26.
[108] S.derin babacan Rafael Molina, Aggelos K.Katsaggelos. Parameter estimationin TV image restoration using variational distribution approximation[J]. IEEETransactions on image processing.200817(3):326-339.
[109] T.E.Bishop, S.D.BaBacan, B.Amizic, A.K.Katsaggelos, Tony Chan andRafeal Molina.“Blind image deconvolution: Problem formulation and existingapproaches,” in blind image deconvolution: Theory and applications, P.Campisi andK.Egiazarian, Eds. Boca Raton, FL:CRC,2007,Ch.1.
[110] D.Mumford and B.Gidas. Stochastic models for generic images[J]. Quart.Appl. Math.,59:85-111,2001.
[111] Tikhonov A.N., Goncharsky A.V., Stepono V.V., et al. Numerical methods forthe solutions of ill-posed problems[M]. Dordrecht: Kluwer Academic Publisher,1995.
[112] Cesari L. On the representation of surfaces[J]. American Journal ofMathematics,1950,72(2):335-346.
[113] Ambrosioa L., Miranda L., Manigliac S., et al. BV functions in abstractWiener spaces[J]. Journal of functional analysis,2010,258(3):785-813.
[114] Ambrosio L., Tortorelli V. M. Approximation of functional depending onjumps by elliptic functional via gamma-convergence[J]. Communication on pureand applied mathematics[J].2006,43(8):999-1036.
[115]吴亚东.图像复原算法研究[D].[博士学位论文].成都：电子科技大学.2006.
[116]冯象初，王卫卫.图像处理的变分和偏微分方程方法[M].科学出版社.2009.
[117] Walter Rudin.实分析与复分析（第三版）[M].戴牧民等译.机械工业出版社.2006.
[118] Candes E. J., Donoho D. L. Curvelets: a surprisingly effective non-adaptiverepresentation for object with edges[R]. Technical report, Stanford: StanfordUniversity,1999.
[119]张军.基于分数阶变分PDE的图像建模与去噪算法研究[D].[博士学位论文].南京：南京理工大学.2009.
[120] Tony F.Chan Jianhong Shen Image Processing and Analysis Variational, PDE,Wavelet and Stochastic Methods[M].科学出版社，2009.
[121] S. Osher, M. Burger, D. Goldfarb, J. Xu, W.Yin. An Iterated regularizationmethod for total variation-based image restoration[J].simal4(2005),pp.460-489.
[122] Y. Wang, J. Yang, W. Yin, Y.Zhang, A new alternating minimization algorithmfor total variation image reconstruction[J], SIAM J. Imag. Sci.,2008.
[123] Gilles Aubert,Pierre Kornprobst. Mathematical problems in image processingpartial differential equations and the calculus of variations(second edition)[M].New York：Springer，2006.
[124] Vogel C.R., Oman M.E. Iterative methods for total variation denoising[J].SIAM Journal on Scientific Computing,1996,17(1):227-238.
[125] T.F.Chan, SELIM ESEDOGLU. Aspects of Total Variation Regularized L1Function Approximation[J]. UCLA CAM Report04-07Feb,2004.
[126] Geman D, Yang C. Nonlinear image recovery with half-quadraticregularization[J]. IEEE Trans. Image Proc,4:932-945,1995.
[127] Katsaggelos A K. Ed.. Digital Image Processing. Berlin:Springer-Verlag.1991.
[128]徐综琦.图像盲复原实用技术研究[D].[硕士学位论文].长沙：国防科学技术大学,2006:45-49.
[129] Charalambous C, Ghaddar F K, Kouris K. Two iterative image restorationalgorithms with application to nuclear medicine[J]. IEEE Trans. on MedicalImaging,11(1):2-8,1992.
[130] Giannis Chantas,Nikolaos Galatsanos,Rafael Molina et al. VariationalBayesian Image Restoration with a Product of Spatially Weighted total VariationImage Priors[J], IEEE Trans. Image Processing,2010,19(2):351-362.
[131]刘伟.基于Non-local means的视频序列去噪[D].[硕士学位论文].西安：西安电子科技大学，2009.
[132]陈前荣.目标图像的光学相关识别与运动模糊图像恢复[D].[博士学位论文].长沙：国防科技大学,2004.
[133]鲍早朝.基于不可分小波和二维主成分分析的图像去噪算法[D].[硕士学位论文].武汉：湖北大学，2008.
[134] Olsen S.I. Estimation of noise in images: An evaluation[J]. Graphical Modelsand Image Processing,1993,55(4):319-323.
[135] Donoho D L, Johnstone IM. Ideal spatial adaptation via wavelet shrinkage[J].Biometrika,1994,81:425-455.
[136] AmerA, Dubois E. Fast and reliable structure-oriented video noiseestimation[J]. IEEE Transactions on Circuits and System for Video Technology,2005,15(1):113-118.
[137] M M Chang, A M Tekalp,and A T Erdem. Blur identification using thebispectrum[J], IEEE Trans. Acoust, Speech, Signal Processing,vol.39,pp.2323-2325,1991.
[138] A E Savakis and H J Trussell, Blur identification by residual spectralmatching [J]. IEEE Trans. Image Process.2(1993),pp.141–151.
[139] Tong,H., Li,M., Zhang,H.J., and Zhang C. Blur detection for digital imagesusing wavelet transform[J]. In Pro. of IEEE Int. Conf. on Multimedia&Expo,2004:17-20.
[140] F.Crete, T.Dolmiere, P.Ladret,and M.Nicolas. The blur effect: perception andestimation with a new no reference perceptual blur metric[J] in Proc. SPIEConference on Human Vision and Electronic Imaging XII,2007,vol.6492.
[141] P.Marziliano, F.Dufaux, S.Winkler and T.Ebrahimi, Perceptual blur andringing metrics: Applications to JPEG2000[J], Signal Proc.:ImageComm.,vol.19,pp.163–172,2004.
[142]罗钧,焦荣梅.一种块效应图像的处理方法[J].计算机技术与发展.2007,17(10):37-39.
[143]张鑫.数字图像的去块效应的研究[D].[硕士学位论文].重庆：重庆大学,2009.
[144] Sheikh H R.Wang and Z, Cormack L.“LIVE image quality assessmentdatabase Release2,” http://live.ece.utexas.edu/research/quality.2009.
[145] H. Paek, R.-C. Kim, and S.-U. Lee, On the POCS-based post processingtechnique to reduce the blocking artifacts in transform coded images[J]. IEEE Trans.Circuits and Systems for Video Tech., vol.8, no.3, pp.358-367, June1998.
[146] Z. Wang, and D. Zhang. A novel approach for the reduction of blockingeffects in low-bit-rate image compression[J]. IEEE Transactions onCommunications, vol.46, no.6, pp.732-734, June1998.
[147] Y. Yang, and N. P. Galatsanos, Removal of compression artifacts usingprojections onto convex sets and line process modeling[J]. IEEE Trans. ImageProcessing, vol.6, no.10.
[148] B. Girod, What’s wrong with mean-squared error[J]. Digital Images andHuman Vision, A. B. Watson, ed., pp.207-220, the MIT press,1993.
[149] S. A. Karunasekera and N. G. Kingsbury, A distortion measure for blockingartifacts in images based on human visual sensitivity[J]. IEEE Trans. ImageProcessing, vol.4, no.6, pp.713-724, June1995.
[150]郭迎春,袁浩杰,吴鹏.一种基于无参考的块效应评价方法[J].光电子激光.22(3):465-468,2011.
[151] Muijs R, Kirenko I. A no-reference blocking artifact measure for adaptivevideoprocessing[J].[EB/OL].http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.127.479&rep=rep1&type=pdf,2005.
[152] Wu H R,Rao K R.Digital video image quality and perceptual coding[M].CRC Press,2006.305-309.
[153] Parvez Sazzad Z M, KawayokeY, HoritaY. No-reference image qualityassessment for JPEG2000based on spatial features[J]. Signal Processing: ImageCommunication,2008,23(4):257-268.
[154]赵巨峰,冯华君等.基于模糊度和噪声水平的图像质量评价方法[J].光电子激光,2010,21(7),1062–1066.
[2] Andrews.H.C, Hunt.B.R. Digital Image Restoration[M]. Englewood Cliffs NewJersey: Prentice-Hall,1977.
[3]邹谋炎.反卷积和信号复原[M].北京,国防工业出版社，2001.4-7
[4] R C Gonzalez, R E Woods.数字图像处理（第二版）[M].阮秋琦等译.北京：电子工业出版社，2007.
[5] Kundur D, Hatzinakos D. Blind image deconvolution revisited[J]. IEEETransactions on signal processing.1996,13(6):61-63.
[6] Q Shan, J Jia, A Agarwala. High-quality motion deblurring from a singleimage[J]. ACM Transaction on Graphics,2008,27(3)：1-10.
[7]邓泽峰，熊有伦.基于频域方法的运动模糊方向识别[J].光电工程.2007,34(10):98-101.
[8] J F Cai, H Ji, C Liu, et al. Blind motion deblurring from a single image usingsparse approximation[C]. IEEE Conference on Computer Vision and PatternRecognition, Miami Beach, FL:IEEE Press,2009:104-111.
[9]李铁成,陶小平,冯华君等.基于倾斜刃边法的调制传递函数计算及图像复原[J].光学学报,2010,30(10):2891-2897.
[10] P J Bex, W Makous. Spatical frequcncy, phase, and the contrast of naturalimages[J]. Journal of the optical society of America,2002,19(6):1096-1106.
[11] J.Da Rugna, H.Konik. Automatic blur detection for metadata extraction incontent-based retrival context[C]. In SPIE, volume5304, pages285-294,2003.
[12] Renting Liu, Zhaorong Li, Jiaya Jia. Image Partial Blur Detection andClassification[C]. CVPR08(1-8). IEEE DOI Link0806.
[13] R.L.Lagendijk, J.Biemod, D.E.Boekee. Identification and restoration of noisyblurred images using the Expectation-Maximization algorithm[J]. IEEETransactions on signal processing.1990,38(7):1180-1190.
[14] R.G. Lane, R.H.T.Bates.Automatic multidimensional deconvolution [J].Journal of the Optical Society.1987,4(1):180-188.
[15] Ayers G.R, Dainty J.C. Iterative blind deconvolution method and itsapplications [J]. Optics Letters,1998,13(7):547-549.
[16] McCallum B C. Bind deconvolution of speckle images[J]. J.Opt.Soc.Am., A.1987,4:180-188.
[17] Rudin L, Osher S, Fatemi E, Nonlinear total variation based noise removalalgorithms[J]. Physica D.1992,60:259-268.
[18] T.F.Chan, C.K.Wong, Total variation blind deconvolution[J].19984(3):370-375.
[19] Esedoglu S, Osher S. Decomposition of images by the anisotropicRudin-Osher-Fatemi model [J]. Communication on Pure and Applied Mathematics.2004,57:1609-1626.
[20] Haiyong Liao, Michael K. Ng. Blind Deconvolution Using GeneralizedCross-Validation Approach to Regularization Parameter Estimation[J]. IEEETransactions on Image Processing,2011,20(3):670-680.
[21] Giannakis Georgios B, Heath Robert W. Blind identification of multi-channelFIR blurs and perfect image restoration[J]. IEEE Transactions on Image Processing,2000,9(11):1877-1896.
[22] J F Cai, H Ji, C Liu, et al. Blind motion deblurring using multiple images[J].Journal of Computational Physics,2009,228(14):5057-5071.
[23] M.Van Noort, L.R. Vaner Voort, M.G.Lofdahl. Solar image restoration by useof multi-frame blind deconvolution with multiple objects and phase diversity[J].Sol. Phys.228,191-215,2005.
[24] A bd-krim Seghouane. A kullback-Leibler divergence approach to blind imagerestoration[J]. IEEE Transactions on Image processing.20(7),2011.
[25] J.H.Money, S.H.Kang. Total variation minimizing blind deconvolution withshock filter reference. Image vis. Comput,26(2):302-314,2008.
[26] L.Bar, N.Sochen, N.Kiryati. Variational pairing of image segmentation andblind restoration[J]. in Proc.8th Eur. Conf. Computer Vision, Prague, Czech Public,May,2004, vol. Part II, LNCS3022:166-177.
[27] M.S.C. Almeida, L.B.Almeida. Blind deblurring of natural images. Presentedat the ICASSP, Las Vegas, NV, Apr,2008.
[28] H. R. Wu, and M. Yuen, A generalized block-edge impairment metric for videocoding[J]. IEEE Signal Proc.Letters, vol.4, no.11, pp.317-320, Nov.1997.
[29] D. Wei and A. C. Bovik, A new metric for blocking artifacts in blocktransform-coded images and video[J]. Technical Report, The University of Texas,Oct.1998.
[30] Hamid Rahim Sheikh, Alan Conrad Bovik. No-Reference Quality AssessmentUsing Natural Scene Statistics: JPEG2000[J]. IEEE Transactions on imageprocessing, vol.14, no.11, nov2005.
[31] Alan C,Bovki, Shizhong Liu. DCT-domain blind measurement of blockingartifacts in DCT-coded images[C]. IEEE International Conference onAcoustics,Speech and Signal Processing. Slat Lake City,2001,1725-1728.
[32]张航，罗大庸.图像盲复原算法研究现状及其展望[J].中国图象图形学报,2004.9(10):1145-1152.
[33] M.Bronstein, A.Bronstein, M.Zibulevsky and Y.Zeevi. Blind deconvolution ofimages using optimal sparse representations[J]. IEEE Transaction on imageprocessing, vol.14, no.6, pp.726-736, Jun.2005.
[34] Deng Zhipo, Xia Youshen. A neural network algorithm for fast blind imagerestoration using a novel2D-ARMA parameter estimation[J]. Internationalconference on Audion Language and Image Processing,2010.
[35] T.A.Cheema, I.M.Qureshi, A.Hussain. Blind image deconvolution usingspace-variant neural network approach[J]. Electronics Letters.41(6):308-309,2005.
[36] Lun Daniel P K, Hsung T C, Shen T W. Orthogonal discrete periodic Radontransform. Part I: Theory and realization[J]. Signal Processing,2003,83(5):941-955.
[37] Lun Daniel P K, Hsung T C, Shen T W. Orthogonal discrete periodic Radontransform. Part II: applications[J]. Signal Processing,2003,83(5):957-971.
[38] Y. Wang, W.Yin, Y.Zhang, A fast algorithm for image deblurring with totalvariation regularization[J]. CAAM Tech. Rep. TR07-10(2007), Rice Univ., Houston,TX,2007.
[39]李仕.航空异速像移模糊实时恢复算法研究与GPU平台实现[D].［博士学位论文］．长春：中国科学院长春光学精密机械与物理研究所,2010.
[40]沈峘,李舜酩,毛建国,辛江慧.数字图像复原技术综述[J].中国图象图形学报.2009.
[41]颜昌翔，王家琪.航空相机像移补偿计算的坐标变换方法[J].光学精密工程，2000，8(3):203-207.
[42] Reeves S I, Mersereau R M, Blur identification by the method of generalizedcross-validation[J]. IEEE Transactions on Image Processing,1992,1(3):301-311.
[43]史光辉.像移对卫星摄影成像质量的影响[J].光学精密工程，1997,5(4):31-34
[44]王晓红,赵荣椿.匀速直线运动模糊的PSF之估计[J].计算机应用.2001,21(9):40-41.
[45]王晓红,赵荣椿.任意方向运动模糊的消除[J].中国图象图形学报.2000,5(6):525-529.
[46]苏光大.图像并行处理技术[M].清华大学出版社.2002,64-65.
[47] Stokseth P.A. Properties of a defocused optical system [C], J. Opt. Soc. Am.1969,59:1314-1321.
[48] J.W.Hardy. Adaptive optics for astronomical telescopes[M]. Oxford UniversityPress, New York,1998.
[49] M.C.Roggemann and B.Welsh. Imagine Through Turblence [M]. CTC Press,Boca Raton, FL,1996.
[50] C.Vogel. Computational Methods for Inverse Problems[J]. AppliedMathematics. SIAM,2002.
[51]杨枝灵，王开. Visual C++数字图形获取处理及实践应用[M].北京：人民邮电出版社，2003.
[52] Sondhi M M, Image restoration: The removal of spatially invariant degrada-tions[J], Proceedings of the IEEE,1972,60(7):842-853.
[53]王旭辉,郭光亚.匀速运动模糊图像的快速恢复[J].微计算机信息.2000,16(5):5-7.
[54] R Hunt B. A Matrix Theory Proof of the Discrete Convolution Theorem[C].IEEE Trans.1971, AU-19(4):285-288.
[55] Jain A K, Fundamentals of Digital Image Processing[M]. Prentice-Hall,Englewood Cliffs, N J.,1989.
[56]贾平,张葆,孙辉.航空成像像移模糊恢复技术[J].光学精密工程,2006,14(4):697-703.
[57]张宏林. Visual C++数字图像处理典型算法及实现[M].人民邮电出版社,2008.
[58]刘微，运动模糊图像恢复算法的研究与实现[D],［博士学位论文］．长春：中国科学院长春光学精密机械与物理研究所,2006.
[59]王新成.高级图像处理技术[M].中国科学技术出版社,2001.
[60] Rafael Molina, Aggelos K. Katsaggelos, Javier Mateos. Bayesian andregularization methods for hyperparameter estimation in image restoration[J]. IEEETransactions on Image Processing.8(2):231-246.
[61] Kulmer S. Information Theory and statistics[M]. New York: Dover.
[62] Portilla J, Simoncelli E. Image restoration using Gaussian scale mixtures in thewavelet domain[A]. Proceedings of the IEEE Conference on Image Processing[C].IEEE Signal Processing Society,2004,2:965-968.
[63] Kundur D. Blind deconvolution of still images using recursive inversefiltering[J]. M.A.Sc. Thesis, University of Toronto, Department of electrical andComputer Engineering.1995,1-19.
[64] Kundur D, Hatzinakos D. A Novel blind deconvolution scheme for imagerestoration using recursive filtering[J]. IEEE Transactions on Signal Proeessing，1998，46(2):375-390.
[65] Kundur D, Hatzinakos D. On the use of lyapunov criteria to analyze theconvergence of blind deconvolution algorithms[J]. IEEE Transactions on SignalProeessing，1998，46(11):2918-2925.
[66]来沛剑，基于全变分的含噪图像盲复原算法研究[D],[硕士学位论文].昆明：昆明理工大学,2010.
[67] Rafael Molina. Javoer Mateos, Aggelos K. Katsaggelos. Blind deconvolutionusing a variational approach to parameter, image and blur estimation[J]. IEEE Trans.Image processing.15(12):3715-3727,2006.
[68] A.K.Katsaggelos, K.T.Lay. Maximum likelihood blur identification and imagerestoration using the EM algorithm[J]. IEEE Trans. signal processing.39(3):729-733,2004.
[69] A.C.Likas, N.P.Galatsanos. A variational approach for Bayesian blind imagedeconvolution[J]. IEEE Trans. signal processing.52(8):2222-2233,2004.
[70] Giannis Chantas, Nikolaos, Aristidis Likas, Michael Saunders. Variationalbayesian image restoration based on a product of t-distributions image prior[J].IEEE Trans. image processing,200817(10):1795-1805.
[71]魏政刚，袁杰辉，蔡元龙.图象质量评价方法的历史、现状和未来[J].中国图象图形学报，1998，3(5):386-389.
[72] Eskicioglu A M, Fisher P S, Image Quality Measures and Their Performance[J].IEEE Transactions on Communications.1995,43.
[73] Wang Z, Cormack L, et al. Laboratory for Image&Video Engineering[EB/OL]. http://live.ece.utexas.edu/research/quality,2002.
[74]林海祥，张炘.无参考图像质量评价综述[J].电脑知识与技术，2009,5(28):8043-8046.
[75] Wang Z, Bovik A C, A Universal Image Quality Index[J], IEEE SIGNALPROCESSING LETTERS.2002,9.
[76] Hosaka K，A New Picture Quality Evaluation Method[C], Proceedings of theInternational Picture Coding Symposium, April1986.
[77] Olsson S, Stropplan A M, Baina J,Objective methods for assessment of videoquality: state of the art [J]. IEEE Transactions on Broadcasting,1997,43(4):487-495
[78] Z.Wang, A.C.Bovik, H.R.Sheikh, et al. Image quality assessment: from errorvisibility to structural similarity. IEEE Transactions on Image Processing.2004,13(4):600-612.
[79]肖文.空域无参考的图像质量评价[D].[硕士学位论文].南昌：江西财经大学，2006.
[80]曹茂永，孙农亮，郁道银.基于灰度梯度的数字图像评价函数[J].光电工程，2003，30(6):69-72.
[81]胡良梅,高隽,何柯峰.图像融合质量评价方法的研究[J].电子学报.2004,32(12):218-221.
[82]丁艺芳.基于小波变换和视觉系统的图像质量综合评价新方法[D].[博士学位论文].上海：上海大学.2001.
[83]徐鲁安,章琦等.一种新的图像质量评价方法[J].计算机工程与设计.2004,25(3):418-420.
[84]刘艳春.利用传播波方程作运动模糊图像质量恢复的客观评价[D].[硕士学位论文].广州：暨南大学.2002.
[85]崔清亮.图像质量评价在虹膜识别中的应用研究[D].[硕士学位论文].长春：吉林大学.2004.
[86]魏政刚，袁杰辉,蔡元龙.一种基于视觉感知的图像质量评价方法[J].电子学报，1999，27(4):79-82.
[87] Osberger W, Maeder A J. A technique for image quality assessment based on ahuman visual systerm model. EUSIPCO-98.
[88]吴振宇，杜少军，姚洪利.采用聚焦度量的散焦模糊图像参数鉴别及复原[J].红外与激光工程，2011，40(4):772-776.
[89] Aihong Liu, Guosheng Hu, Haiyan Wu. A New Method for Blur Identificationof Out-of-Focus Images[J]. Education Technology and Computer Science,2009,298-303.
[90] Cannon.TM. Blind deconvolution of spatially invariant blurs with phase[C].IEEE Trans. on Acoustics Speech and signal processing,1976,24(1):58-63.
[91]穆丽娟.数字图像自动聚焦与离焦模糊复原技术研究[D].[硕士学位论文].天津：天津师范大学.2009.
[92]赵琳,金伟其,陈翼男等.基于微分图像自相关的离焦模糊图像盲复原[J].光学学报,2008,28,(9):1703-1709.
[93]陈前荣等.利用拉氏算子鉴别散焦模糊图像点扩散函数[J].计算机工程与科学.2005,27(9):40-43
[94]赵书安等. Hough变换在几何特征检测中的应用[J].南京师范大学学报（工程技术版）.2006,6(4):66-70.
[95]朱芳芳等.一种改进的Hough变换直线检测算法[J].计算机技术与发展.2009,19(5):19-22.
[96]刘宏申，程健.广义Hough变换算法的分析改进[J].中国科学技术大学学报.2009,39(11):1202-1206.
[97]杨全银，魏雅杰，王欣.基于Hough变换的线段检测算法[J].红外与激光工程，2009,38(3):525-529.
[98]秦开怀，王海颖，郑辑涛.一种基于Hough变换的圆和矩形的快速检测方法[J].中国图象图形学报.2010,15(1):109-115.
[99]杨全银.基于Hough变换的图像形状特征检测[D].[硕士学位论文].济南：山东大学.2009.
[100]王彬生，黄乡生. Hough变换及其在几何特征检测中的应用[J].计算机与现代化.2008,4:20-22.
[101] Tikhonov A N, Arsenin V Y. Solutions of ill-posed problem. New York:Wiley,1977:87-94
[102] Hadamard J. Lectures on the Cauchy problems in linear partial differentialequations. Yale University Press, New Haven.1923.
[103] Kirsch A. An Introduction to the Mathematical Theory of Inverse Problems.NewYork: Springer-Verlag,1996:5-11.
[104] R. L. Lagendijk, J. Biemond, D. E. Boekee. Regularized iterative imagerestoration with ringing reduction[J]. IEEE Transactions on Acoustics, Speech, andSignal Processing,1988,36(12):1874-1887.
[105]郭永彩,王婀娜,高潮.空间自适应和正则化技术的盲图像复原[J].光学精密工程，2008,16(11):2263-2267.
[106] Song H H, Zhang L, Wang P K, et al. An adaptive L1-L2hybrid error modelto superresolution[C]. Image Processing, IEEE International Conference,2010.2812-2824.
[107] S.Derin Babacan, Rafael Molina, Aggelos K.Katsaggelos. Variationalbayesian blind deconvolution using a total variation prior[J]. IEEE Transactions onimage processing,2009,18(1):12-26.
[108] S.derin babacan Rafael Molina, Aggelos K.Katsaggelos. Parameter estimationin TV image restoration using variational distribution approximation[J]. IEEETransactions on image processing.200817(3):326-339.
[109] T.E.Bishop, S.D.BaBacan, B.Amizic, A.K.Katsaggelos, Tony Chan andRafeal Molina.“Blind image deconvolution: Problem formulation and existingapproaches,” in blind image deconvolution: Theory and applications, P.Campisi andK.Egiazarian, Eds. Boca Raton, FL:CRC,2007,Ch.1.
[110] D.Mumford and B.Gidas. Stochastic models for generic images[J]. Quart.Appl. Math.,59:85-111,2001.
[111] Tikhonov A.N., Goncharsky A.V., Stepono V.V., et al. Numerical methods forthe solutions of ill-posed problems[M]. Dordrecht: Kluwer Academic Publisher,1995.
[112] Cesari L. On the representation of surfaces[J]. American Journal ofMathematics,1950,72(2):335-346.
[113] Ambrosioa L., Miranda L., Manigliac S., et al. BV functions in abstractWiener spaces[J]. Journal of functional analysis,2010,258(3):785-813.
[114] Ambrosio L., Tortorelli V. M. Approximation of functional depending onjumps by elliptic functional via gamma-convergence[J]. Communication on pureand applied mathematics[J].2006,43(8):999-1036.
[115]吴亚东.图像复原算法研究[D].[博士学位论文].成都：电子科技大学.2006.
[116]冯象初，王卫卫.图像处理的变分和偏微分方程方法[M].科学出版社.2009.
[117] Walter Rudin.实分析与复分析（第三版）[M].戴牧民等译.机械工业出版社.2006.
[118] Candes E. J., Donoho D. L. Curvelets: a surprisingly effective non-adaptiverepresentation for object with edges[R]. Technical report, Stanford: StanfordUniversity,1999.
[119]张军.基于分数阶变分PDE的图像建模与去噪算法研究[D].[博士学位论文].南京：南京理工大学.2009.
[120] Tony F.Chan Jianhong Shen Image Processing and Analysis Variational, PDE,Wavelet and Stochastic Methods[M].科学出版社，2009.
[121] S. Osher, M. Burger, D. Goldfarb, J. Xu, W.Yin. An Iterated regularizationmethod for total variation-based image restoration[J].simal4(2005),pp.460-489.
[122] Y. Wang, J. Yang, W. Yin, Y.Zhang, A new alternating minimization algorithmfor total variation image reconstruction[J], SIAM J. Imag. Sci.,2008.
[123] Gilles Aubert,Pierre Kornprobst. Mathematical problems in image processingpartial differential equations and the calculus of variations(second edition)[M].New York：Springer，2006.
[124] Vogel C.R., Oman M.E. Iterative methods for total variation denoising[J].SIAM Journal on Scientific Computing,1996,17(1):227-238.
[125] T.F.Chan, SELIM ESEDOGLU. Aspects of Total Variation Regularized L1Function Approximation[J]. UCLA CAM Report04-07Feb,2004.
[126] Geman D, Yang C. Nonlinear image recovery with half-quadraticregularization[J]. IEEE Trans. Image Proc,4:932-945,1995.
[127] Katsaggelos A K. Ed.. Digital Image Processing. Berlin:Springer-Verlag.1991.
[128]徐综琦.图像盲复原实用技术研究[D].[硕士学位论文].长沙：国防科学技术大学,2006:45-49.
[129] Charalambous C, Ghaddar F K, Kouris K. Two iterative image restorationalgorithms with application to nuclear medicine[J]. IEEE Trans. on MedicalImaging,11(1):2-8,1992.
[130] Giannis Chantas,Nikolaos Galatsanos,Rafael Molina et al. VariationalBayesian Image Restoration with a Product of Spatially Weighted total VariationImage Priors[J], IEEE Trans. Image Processing,2010,19(2):351-362.
[131]刘伟.基于Non-local means的视频序列去噪[D].[硕士学位论文].西安：西安电子科技大学，2009.
[132]陈前荣.目标图像的光学相关识别与运动模糊图像恢复[D].[博士学位论文].长沙：国防科技大学,2004.
[133]鲍早朝.基于不可分小波和二维主成分分析的图像去噪算法[D].[硕士学位论文].武汉：湖北大学，2008.
[134] Olsen S.I. Estimation of noise in images: An evaluation[J]. Graphical Modelsand Image Processing,1993,55(4):319-323.
[135] Donoho D L, Johnstone IM. Ideal spatial adaptation via wavelet shrinkage[J].Biometrika,1994,81:425-455.
[136] AmerA, Dubois E. Fast and reliable structure-oriented video noiseestimation[J]. IEEE Transactions on Circuits and System for Video Technology,2005,15(1):113-118.
[137] M M Chang, A M Tekalp,and A T Erdem. Blur identification using thebispectrum[J], IEEE Trans. Acoust, Speech, Signal Processing,vol.39,pp.2323-2325,1991.
[138] A E Savakis and H J Trussell, Blur identification by residual spectralmatching [J]. IEEE Trans. Image Process.2(1993),pp.141–151.
[139] Tong,H., Li,M., Zhang,H.J., and Zhang C. Blur detection for digital imagesusing wavelet transform[J]. In Pro. of IEEE Int. Conf. on Multimedia&Expo,2004:17-20.
[140] F.Crete, T.Dolmiere, P.Ladret,and M.Nicolas. The blur effect: perception andestimation with a new no reference perceptual blur metric[J] in Proc. SPIEConference on Human Vision and Electronic Imaging XII,2007,vol.6492.
[141] P.Marziliano, F.Dufaux, S.Winkler and T.Ebrahimi, Perceptual blur andringing metrics: Applications to JPEG2000[J], Signal Proc.:ImageComm.,vol.19,pp.163–172,2004.
[142]罗钧,焦荣梅.一种块效应图像的处理方法[J].计算机技术与发展.2007,17(10):37-39.
[143]张鑫.数字图像的去块效应的研究[D].[硕士学位论文].重庆：重庆大学,2009.
[144] Sheikh H R.Wang and Z, Cormack L.“LIVE image quality assessmentdatabase Release2,” http://live.ece.utexas.edu/research/quality.2009.
[145] H. Paek, R.-C. Kim, and S.-U. Lee, On the POCS-based post processingtechnique to reduce the blocking artifacts in transform coded images[J]. IEEE Trans.Circuits and Systems for Video Tech., vol.8, no.3, pp.358-367, June1998.
[146] Z. Wang, and D. Zhang. A novel approach for the reduction of blockingeffects in low-bit-rate image compression[J]. IEEE Transactions onCommunications, vol.46, no.6, pp.732-734, June1998.
[147] Y. Yang, and N. P. Galatsanos, Removal of compression artifacts usingprojections onto convex sets and line process modeling[J]. IEEE Trans. ImageProcessing, vol.6, no.10.
[148] B. Girod, What’s wrong with mean-squared error[J]. Digital Images andHuman Vision, A. B. Watson, ed., pp.207-220, the MIT press,1993.
[149] S. A. Karunasekera and N. G. Kingsbury, A distortion measure for blockingartifacts in images based on human visual sensitivity[J]. IEEE Trans. ImageProcessing, vol.4, no.6, pp.713-724, June1995.
[150]郭迎春,袁浩杰,吴鹏.一种基于无参考的块效应评价方法[J].光电子激光.22(3):465-468,2011.
[151] Muijs R, Kirenko I. A no-reference blocking artifact measure for adaptivevideoprocessing[J].[EB/OL].http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.127.479&rep=rep1&type=pdf,2005.
[152] Wu H R,Rao K R.Digital video image quality and perceptual coding[M].CRC Press,2006.305-309.
[153] Parvez Sazzad Z M, KawayokeY, HoritaY. No-reference image qualityassessment for JPEG2000based on spatial features[J]. Signal Processing: ImageCommunication,2008,23(4):257-268.
[154]赵巨峰,冯华君等.基于模糊度和噪声水平的图像质量评价方法[J].光电子激光,2010,21(7),1062–1066.