盲均衡技术在医学CT图像盲恢复算法中的应用研究
|
摘要
医学CT图像作为进行疾病检查和诊断的重要依据，其质量好坏直接影响诊断的准确性。在医学CT成像过程中不可避免地受到点扩展函数的影响，使得图像产生退化，影响诊断效果，且退化过程往往是未知的。图像盲恢复算法是在未知图像退化过程的前提下，仅利用退化图像来消除点扩展函数影响的图像恢复技术，广泛应用于天文成像、医疗诊断、军事公安等领域。盲均衡技术目前已扩展应用到图像盲恢复中，它是将点扩展函数影响的消除等效为码间干扰的消除，已成为通信信号处理与图像分析相结合的前沿热点研究课题。
本文所做的主要工作如下：
（1）提出了降维处理的医学CT图像盲均衡算法，构建了应用于医学CT图像盲均衡算法的代价函数，推导了算法迭代公式，分析了算法收敛性能，进行了计算机仿真；其次，利用变步长思想解决了算法收敛速度和收敛精度之间的矛盾，提出了基于误差信号峰度和均方误差变步长的医学CT图像盲均衡算法，仿真表明新算法具有较快的收敛速度和较小的稳态误差；第三，通过在权值迭代中引入代价函数的二阶Hessian矩阵，提出了分数低阶恒模医学CT图像盲均衡算法，改善了算法的收敛性能，提高了峰值信噪比。
（2）提出了行列变换的恒模医学CT图像盲均衡算法，通过行列变换将点扩展函数的影响分解为横纵两个方向，利用复值系统盲均衡算法在I和Q方向彼此不影响的特性，分别进行点扩展函数影响的消除，推导了算法迭代公式，分析了算法稳态和动态收敛性能，计算机仿真表明，新算法提高了峰值信噪比和改善信噪比。同时通过频域变换，提出了频域差错概率最小医学CT图像盲均衡算法，分析了迭代步长选取原则，进行了计算机仿真，验证了算法的有效性。
（3）提出了基于Zigzag编码和双Zigzag编码两种医学CT图像神经网络盲均衡算法，选取三层前馈神经网络结构，设计了传递函数，推导了迭代公式，分析了收敛性能，并进行了计算机仿真。实验结果表明，新算法改善了恢复效果，降低了均方误差，提高了峰值信噪比和改善信噪比。
Medical CT image is one of main basis for the diagnosis and treatment ofdisease, and its quality will directly affects the accuracy of diagnosis. However, in theprocess of medical CT imaging, due to the impact of the point spread function, imagewill emerge from degradation, and affect diagnosis effect. And the exact cause ofdegradation may be unknown. Blind image restoration algorithm is that the unknownimage, blur and all model parameters, including the noise variances, are estimatedsolely from the observations without prior knowledge or user intervention. It waswidely utilized in the field of astronomical imaging, medical diagnostics, military andpublic security. Preliminary results of blind equalization algorithm were first found inthe signal case then extended to the image. Reducing the affect of point spreadfunction is equivalent to eliminating the inter-symbol interference. It has become ahot research topic in the field of communication signal processing and imageanalysis.
The major contribution of this paper is summarized as follow:
（1） A medical CT image blind equalization algorithm based on dimensionreduction was proposed in this paper. We defined the cost function applied to medicalCT image and demonstrate the performance of convergence. At the same time, thestrategy of variable step size was utilized to solve the contradiction betweenconvergence speed and accuracy. Two medical CT image blind equalizationalgorithms based on reduction dimension were proposed. Simulation results show thatthe proposed algorithms have faster convergence rate and smaller steady state error.In order to speed up the convergence of constant modulus algorithm and improve theperformance of algorithm, the second rank Hessian information of cost function wasutilized in the process of weight update. Computer simulations demonstrate the newalgorithm improves the convergence of the algorithm performance and peak signalnoise ratio.
（2） A medical CT image constant module blind equalization algorithm basedon row-column transform was proposed. The affect of point spread function was decomposed to vertical and horizontal direction. The characteristics that I and Qdirection does not affect each other in the complex blind equalization algorithm, wasutilized to eliminate the affect of point spread function. Iteration formula was derivedand the static and dynamical convergence performance was analyzed. It is shown in aseries of computer-simulated experiments that the proposed method outperforms anumber of existing alternatives in terms of peak signal to noise ratio and recoveryeffects. By use of frequency domain transform, the frequency domain minimum errorprobability medical CT image blind equalization algorithm was proposed, theselection principle of the step size was analyzed, and computer simulation verify thevalidity of the algorithm.
（3） Both medical CT image neural network blind equalization algorithmsbased on Zigzag coding and double Zigzag coding were proposed. Three layer neuralnetwork structures were adopted. We designed the transfer function of neural network,derived iteration formula and analyzed convergence performance. Computersimulation experiments show that the proposed algorithm reduces mean square errorand improves restoration effect, peak signal to noise ratio and improving signal tonoise ratio.
本文所做的主要工作如下：
（1）提出了降维处理的医学CT图像盲均衡算法，构建了应用于医学CT图像盲均衡算法的代价函数，推导了算法迭代公式，分析了算法收敛性能，进行了计算机仿真；其次，利用变步长思想解决了算法收敛速度和收敛精度之间的矛盾，提出了基于误差信号峰度和均方误差变步长的医学CT图像盲均衡算法，仿真表明新算法具有较快的收敛速度和较小的稳态误差；第三，通过在权值迭代中引入代价函数的二阶Hessian矩阵，提出了分数低阶恒模医学CT图像盲均衡算法，改善了算法的收敛性能，提高了峰值信噪比。
（2）提出了行列变换的恒模医学CT图像盲均衡算法，通过行列变换将点扩展函数的影响分解为横纵两个方向，利用复值系统盲均衡算法在I和Q方向彼此不影响的特性，分别进行点扩展函数影响的消除，推导了算法迭代公式，分析了算法稳态和动态收敛性能，计算机仿真表明，新算法提高了峰值信噪比和改善信噪比。同时通过频域变换，提出了频域差错概率最小医学CT图像盲均衡算法，分析了迭代步长选取原则，进行了计算机仿真，验证了算法的有效性。
（3）提出了基于Zigzag编码和双Zigzag编码两种医学CT图像神经网络盲均衡算法，选取三层前馈神经网络结构，设计了传递函数，推导了迭代公式，分析了收敛性能，并进行了计算机仿真。实验结果表明，新算法改善了恢复效果，降低了均方误差，提高了峰值信噪比和改善信噪比。
Medical CT image is one of main basis for the diagnosis and treatment ofdisease, and its quality will directly affects the accuracy of diagnosis. However, in theprocess of medical CT imaging, due to the impact of the point spread function, imagewill emerge from degradation, and affect diagnosis effect. And the exact cause ofdegradation may be unknown. Blind image restoration algorithm is that the unknownimage, blur and all model parameters, including the noise variances, are estimatedsolely from the observations without prior knowledge or user intervention. It waswidely utilized in the field of astronomical imaging, medical diagnostics, military andpublic security. Preliminary results of blind equalization algorithm were first found inthe signal case then extended to the image. Reducing the affect of point spreadfunction is equivalent to eliminating the inter-symbol interference. It has become ahot research topic in the field of communication signal processing and imageanalysis.
The major contribution of this paper is summarized as follow:
（1） A medical CT image blind equalization algorithm based on dimensionreduction was proposed in this paper. We defined the cost function applied to medicalCT image and demonstrate the performance of convergence. At the same time, thestrategy of variable step size was utilized to solve the contradiction betweenconvergence speed and accuracy. Two medical CT image blind equalizationalgorithms based on reduction dimension were proposed. Simulation results show thatthe proposed algorithms have faster convergence rate and smaller steady state error.In order to speed up the convergence of constant modulus algorithm and improve theperformance of algorithm, the second rank Hessian information of cost function wasutilized in the process of weight update. Computer simulations demonstrate the newalgorithm improves the convergence of the algorithm performance and peak signalnoise ratio.
（2） A medical CT image constant module blind equalization algorithm basedon row-column transform was proposed. The affect of point spread function was decomposed to vertical and horizontal direction. The characteristics that I and Qdirection does not affect each other in the complex blind equalization algorithm, wasutilized to eliminate the affect of point spread function. Iteration formula was derivedand the static and dynamical convergence performance was analyzed. It is shown in aseries of computer-simulated experiments that the proposed method outperforms anumber of existing alternatives in terms of peak signal to noise ratio and recoveryeffects. By use of frequency domain transform, the frequency domain minimum errorprobability medical CT image blind equalization algorithm was proposed, theselection principle of the step size was analyzed, and computer simulation verify thevalidity of the algorithm.
（3） Both medical CT image neural network blind equalization algorithmsbased on Zigzag coding and double Zigzag coding were proposed. Three layer neuralnetwork structures were adopted. We designed the transfer function of neural network,derived iteration formula and analyzed convergence performance. Computersimulation experiments show that the proposed algorithm reduces mean square errorand improves restoration effect, peak signal to noise ratio and improving signal tonoise ratio.
引文
[01] Sato Y， A Method of Self-Recovering Equalization for MultilevelAmplitude-Modulation Systems，IEEE Transactions on Communication，1975，23（6）：679~682
[02]刘涛，郭军，盲图像恢复中的二维盲均衡研究，电子与信息学报，2006，28（6）：1013~1015
[03]张金玉，黄先祥，谢伟达，基于盲均衡理论的弱冲击故障的检测研究，机械科学与技术，2008，27（8）：996~999
[04] Xiaohua Li，Blind Channel Estimation and Equalization in Wireless SensorNetworks Based on Correlations among Sensors，IEEE Transactions on SignalProcessing，2005，53（4）：1511~1519
[05] Bruneau M，Reinhorn A.M， Constantinou M.C， et al，The University atBuffalo（UB）Node of the NEES Network–a Versatile High Performance TestingFacility towards Real-time Dynamic Hybrid Testing，12th European Conferenceon Earthquake Engineering London，UK，2002：1~8
[06] Pavlovic G， Tekalp A.M， Maximum Likelihood Parametric based on aContinuous Spatial Blur Identification Domain Model，IEEE Transactions onImage Processing，1992，I（4）：496~504
[07] Li D，Mersereau R.M，Simske S，Blind Image Deconvolution Through SupportVector Regression，IEEE Transactions on Neural Networks，2007，18（3）：931~935
[08] Jingyan Xu，Taguchi K，Tsui B.M.W，Statistical Projection Completion in X-rayCT Using Consistency Conditions，IEEE Transactions on Medical Imaging，2010，29（8）：1528~1540
[09] Figueiredo M.A.T，Bioucas-Dias J.M，Restoration of Poissonian Images usingAlternating Direction Optimization，IEEE Transactions on Image Processing，2010，19（12）：3133~3145
[10] Reeves S.J，Mersereau R.M，Blur Identification by the Method of GeneralizedCross-validation，IEEE Transaction on Image Processing，1992，1（3）：301~311
[11] Nguyen N，Golub G，Milanfar P，Blind restoration/superresolution withgeneralized cross-validation using Gauss-type quadrature rules，Record of theThirty-Third Asilomar Conference on Signals, Systems, and Computers，PacificGrove, CA, USA，1999，（2）：1257~1261
[12] Haiyong Liao， Ng M.K， Blind Deconvolution Using GeneralizedCross-Validation Approach to Regularization Parameter Estimation， IEEETransactions on Image Processing，2011，20（3）：670~680
[13] Tsumuraya F，Miura N，Baba N，Iterative Blind Deconvolution Method usingLucy’s Algorithm，Astronomy and Astrophysics，1994，282：669~708
[14] Likas A.C，Galatsanos N.P，A Variational Approach for Bayesian Blind ImageDeconvolution，IEEE Transactions on Signal Processing，2004，52（8）：2222~2233
[15] Gupta S，Chauhan R.C，Saxena S.C，Locally Adaptive Wavelet DomainBayesian Processor for Denoising Medical Ultrasound Images using SpeckleModelling based on Rayleigh Distribution，IEE Proceeding of Vision, Image andSignal Processing，2005，152（1）：129~135
[16] Tzikas D.G， Likas A.C， Galatsanos N.P， Variational Bayesian SparseKernel-Based Blind Image Deconvolution With Student’s-t Priors， IEEETransactions on Image Processing，2009，18（4）：753~764
[17] Bishop T.E，Molina R，Hopgood J.R，Blind Restoration of Blurred Photographsvia AR Modelling and MCMC，15th IEEE International Conference on ImageProcessing，San Diego, CA, United states，2008：669~672
[18] Min Dai，Cheng Peng，Chan A.K，et al，Bayesian Wavelet Shrinkage with EdgeDetection for SAR Image Despeckling，IEEE Transactions on Geoscience andRemote Sensing，2004，42（8）：1642~1648
[19] Babacan S.D，Jingnan Wang，Molina R，et al，Bayesian Blind DeconvolutionFrom Differently Exposed Image Pairs， IEEE Transactions on ImageProcessing，2010，19（11）：2874~2888
[20] Karayiannis N.B，Venetsanopoulos A.N，Regularization Theory in ImageRestoration-The Stabilizing Functional Approach， IEEE Transactions onAcoustics, Speech, and Signal Processing，1990，38（7）：1155~1179
[21] Kim-Hui Yap，Ling Guan，A Computational Reinforced Learning Scheme toBlind Image Deconvolution，IEEE Transactions on Evolutionary Computation，2002，6（1）：2~15
[22] Michailovich O.V，Adam D，A Novel Approach to the2-D Blind DeconvolutionProblem in Medical Ultrasound，IEEE Transactions on Medical Imaging，2005，24（1）：86~104
[23] Belekos S.P，Galatsanos N.P，Katsaggelos A.K，Maximum a Posteriori VideoSuper-Resolution Using a New Multichannel Image Prior，IEEE Transactions onImage Processing，2010，19（6）：1451~1464
[24] Humphrey D，Taubman D，A Filtering Approach to Edge Preserving MAPEstimation of Images，IEEE Transactions on Image Processing，2011，20（5）：1234~1248
[25] Ming Jiang，Ge Wang，Skinner，M.W，et al，Blind Deblurring of Spiral CTImages，IEEE Transactions on Medical Imaging，2003，22（7）：837~845
[26] Li Chen，Kim-Hui Yap，A Soft Double Regularization Approach to ParametricBlind Image Deconvolution，IEEE Transactions on Image Processing，2005，14（5）：624~633
[27] Li Chen，Kim-Hui Yap，Efficient Discrete Spatial Techniques for Blur SupportIdentification in Blind Image Deconvolution，IEEE Transactions on SignalProcessing，2006，54（4）：1557~1562
[28] Sanches J.M，Nascimento J.C，Marques J.S，Medical Image Noise Reductionusing the Sylvester–Lyapunov Equation， IEEE Transactions on ImageProcessing，2008，17（9）：1522~1539
[29] Ayers G.R， Dainty J.C， Iterative Blind Deconvolution Method and ItsApplications，Optics Letters，1988，13（7）：547~549
[30] Lam E.Y， Blind Bi-Level Image Restoration with Iterated QuadraticProgramming，IEEE Transactions on Circuits and Systems-II: Express Briefs，2007，54（1）：52~56
[31] Szolgay D，Szirányi T，Optimal Stopping Condition for Iterative ImageDeconvolution by New Orthogonality Criterion，Electronics Letters，2011，47（7）：442~444
[32] Kundur D，Hatzinakos D，A Novel Blind Deconvolution Scheme for BlindImage Restoration using Recursive Filtering，IEEE Transactions on ImageProcessing，1998，46（2）：375~390
[33]薛梅，杨绿溪，邹采荣，用于含噪二值图像的改进NAS-RIF图像盲复原算法，数据采集与处理，2002，17（2）：156~160
[34] Hongzhi Wang，Sun Qi，Yuanyuan Liu，An Improved NAS-RIF Algorithmbased on the Lifting Scheme for Blind Image Restoration， Pacific-AsiaConference on Circuits, Communications and Systems，Chengdu，China，2009：423~426
[35] Chunpeng Song，Hongzhi Wang，Liliang He，et al，An Improved NAS-RIF BlindImage Restoration based on Higher Order Statistics，20068th InternationalConference on Signal Processing，Guilin，China，2006，（4）：553~556
[36] Kundur D，Hatzinakos D，Leung H，Robust Classification of Blurred Imagery，IEEE Transactions on Image Processing，2000，9（2）：243~255
[37] Kundur D，Hatzinakos D，On the Use of Lyapunov Criteria to Analyze theConvergence of Blind Deconvolution Algorithms，IEEE Transactions on SignalProcessing，1998，46（11）：2918~2925
[38] Chin Ann Ong，Chambers J.A，An Enhanced NAS-RIF Algorithm for BlindImage Deconvolution，IEEE Transactions on image processing，1999，8（7）：988~992
[39] Matsuyama M，Tanji Y，Tanaka M，Enhancing the Ability of NAS-RIFAlgorithm for Blind Image Deconvolution，ISCAS2000IEEE InternationalSymposium on Circuit and Systems，Geneva, Switzerland，2000，（4）：553~556
[40] McCallum B.C， Blind Deconvolution by Simulated Annealing， OpticsCommunications，1990，75（2）：101~105
[41] Leipo Yan，Lipo Wang，Image Restoration using Chaotic Simulated Annealing，Proceedings of the International Joint Conference on Neural Networks，Portland,OR, United states，2003，4：3060-3064
[42]张红英，彭启琮，一种改进的模拟退火图像盲复原算法，电子科技大学学报，2008，35（5）：767~769，787
[43] Vogel C.R，Oman M.E，Fast Robust Total Variation Based Reconstruction ofNoisy Blurred Images，IEEE Transactions on Image Processing，1998，7（6）：813~824
[44] Chan T.F，Chiu-Kwong Wong，Total Variation Blind Image Deconvolution，IEEE Transactions on Image Processing，1998，7（3）：370~375
[45] Hongwei Zheng，Hellwich O，An Edge-Driven Total Variation Approach toImage Deblurring and Denoising， Proceedings of the First InternationalConference on Innovative Computing, Information and Control，Beijing，China，2006，（2）：705~710
[46]张航，罗大庸，一种改进的全变差盲图像复原方法，电子学报，2005，33（7）：1288~1290
[47] Chan R.H，Yiqiu Dong，Hintermuller M，An Efficient Two-Phase L1-TV Methodfor Restoring Blurred Images with Impulse Noise，IEEE Transactions on ImageProcessing，2011，19（7）：1731~1739
[48] Michailovich O.V，An Iterative Shrinkage Approach to Total-Variation ImageRestoration，IEEE Transactions on Image Processing，2011，20（5）：1281~1299
[49] Wang S，Liu Z.W，Dong W.S，et al，Total Variation based Image Deblurringwith Nonlocal Self-similarity Constraint，Electronics Letters，2011，47（16）：916~918
[50] Kim-Hui Yap，Ling Guan，Wanquan Liu，A Recursive Soft-Decision Approachto Blind Image Deconvolution，IEEE Transactions on Signal Processing，2003，51（2）：515~526
[51] Cheema T.A，Qureshi I.M，Hussain A，Blind Image Deconvolution usingSpace-variant Neural Network Approach，Electronics Letters，2005，41（6）：308~309
[52] Russo F，Noise Removal from Image Data Using Recursive Neurofuzzy Filters，IEEE Transactions on Instrumentation and Measurement，2000，49（2）：307~314
[53] Senel H.G，Peters R.A II，Dawant B，Topological Median Filters，IEEETransactions on Image Processing，2002，11（2）：89~104
[54] Van De Ville D，Nachtegael M，Van der Weken D，et al，Noise Reduction byFuzzy Image Filtering，IEEE Transactions on Fuzzy Systems，2003，11（4）：429~436
[55] Haixiang Xu，Guangxi Zhu，Fuyuan Peng，et al，Adaptive Fuzzy Switching Filterfor Images Corrupted by Impulse Noise，2004International Conference onCommunications, Circuits and Systems，2004，（2）：792~795
[56]屈志毅，沃炎，任志宏，基于交替迭代和神经网络的盲目图像恢复，计算机学报，2000，23（4）：410~413
[57] Adam D，Michailovich O，Blind Deconvolution of Ultrasound Sequences UsingNonparametric Local Polynomial Estimates of the Pulse，IEEE Transactions onBiomedical Engineering，2002，49（2）：118~131
[58] Wan S，Raju, B.I，Srinivasan M.A，Robust Deconvolution of High-FrequencyUltrasound Images Using Higher-Order Spectral Analysis and Wavelets，IEEETransactions on Ultrasonics, Ferroelectrics, and Frequency Control，2003，50（10）：1286~1295
[59] Yew Kun Chee，Yeng Chai Soh，A Robust Kalman Filter Design for ImageRestoration，Proceedings of Acoustics, Speech, and Signal Processing，SaltLake，UT，United states，2001，（3）：1825~1828
[60] Jing Wang，Ge Wang，Ming Jiang，Blind deblurring of Spiral CT Images basedon ENR and Wiener filter，Journal of X-Ray Science and Technology，2005，13：49~60
[61] Lane R.G，Bates R.H.T，Automatic Multidimensional Deconvolution，Journal of Optical Society of America，1987，4（1）：180~188
[62] Ghigila D.C， Romero L.A， Mastin G.A， Systematic Approach toTwo-dimensional Blind Image Deconvolution by Zero-sheets Separation，Journal of Optical Society of America，1993，10（5）：1024~1036
[63] Nakagaki R， Katsaggelos A.K， A VQ-Based Blind Image RestorationAlgorithm，IEEE Transactions on Image Processing，2003，12（9）：1044~1053
[64] Tsung-Ying Sun，Chan-Cheng Liu，Yu-Peng Jheng，et al，Blind ImageDeconvolution via Particle Swarm Optimization with Entropy Evaluation，Eighth International Conference on Intelligent Systems Design andApplications，Kaohsiung，Taiwan，2008：265~270
[65] Johnson Eric G，Abusshagur Mustafa A G，Image Deconvolution using a MicroGenetic Algorithm，Optics Communications，1997，140（1-3）：6~10
[66] Yang-Chih Lai，Chih-Li Huo，Yu-Hsiang Yu，et al，PSO-based Estimation forGaussian Blur in Blind Image Deconvolution Problem，2011IEEE InternationalConference on Fuzzy Systems，Taipei，Taiwan，2011：1143~1148
[67]张煜东，吴乐南，改进免疫算法用于图像复原，光学精密工程，2009，17（2）：417~425
[68] Umeyama S， Blind Deconvolution of Images using Gabor Filters andIndependent Component Analysis，4th International Symposium on IndependentComponent Analysis and Blind Signal Separation，Nara，Japan，2003：319~324
[69] Dylov D. V，Waller L，Fleischer J. W，Nonlinear Restoration of Diffused Imagesvia Seeded Instability，IEEE Journal of Selected Topics in Quantum Electronics，2012，18（2）：916~925
[70] Gazzah H，Regalia P.A，Delmas J.P，et al，A Blind Multichannel IdentificationAlgorithm Robust to Order Overestimation，IEEE Transactions on SignalProcessing，2002，50（6）：1449~1458
[71] Hendekli C.M， Ertuzun, A， Design of a Multichannel Two-dimensionalDelta-domain Lattice Filter for Noise Removal，IEEE Transactions on SignalProcessing，2001，49（7）：1581~1593
[72] Vrhel M.J， Unser M， Multichannel Restoration with Limited a PrioriInformation，IEEE Transactions on Image Processing，1999，8（4）：527~536
[73] Elad M，Feuer A，Restoration of a Single Superresolution Image from SeveralBlurred, Noisy and Undersampled Measured Images，IEEE Transactions onImage Processing，1997，6（12）：1646~1658
[74] Miura N，Baba N，Extended-object Reconstruction with Sequential use of theIterative Blind Deconvolution Method，Optics Communications，1992，89（5-6）：375~379
[75] Schulz T.J，Multiframe Blind Deconvolution of Astronomical Images，Journal ofthe Optical Society of America A: Optics and Image Science and Vision，1993，10（5）：1064~1073
[76] Moon Gi Kang，Katsaggelos A.K，Simultaneous Multichannel Image Restorationand Estimation of the Regularization Parameters，IEEE Transactions on ImageProcessing，1997，6（5）：774~778
[77] Hung-Ta Pai，Bovik A.C，On Eigenstructure-based Direct Multichannel BlindImage Restoration，IEEE Transactions on Image Processing，2001，10（10）：1434~1446
[78] Sroubek F，Flusser J，Multichannel Blind Iterative Image Restoration，IEEETransactions on Image Processing，2003，12（9）：1094~1106
[79] Giannakis G.B，Heath R.W Jr，Blind Identification of Multichannel FIR Blursand Perfect Image Restoration，IEEE Transactions on Image Processing，2000，9（11）：1877~1896
[80] Hung-Ta Pai，Bovik A.C，Exact Multichannel Blind Image Restoration，IEEESignal Processing Letters，1997，4（8）：217~220
[81] Chow T.W.S，Xiao-Dong Li，Kai-Tat Ng，Double-regularization Approach forBlind Restoration of Multichannel Imagery，IEEE Transactions on Circuits andSystems I: Fundamental Theory and Applications，2001，48（9）：1075~1085
[82] Panci G， Campisi P， Colonnese S， et al， Multichannel Blind ImageDeconvolution Using the Bussgang Algorithm: Spatial and MultiresolutionApproaches，IEEE Transactions on Image Processing，2003，1（211）：1324~1337
[83] Souidene W，Abed-Meraim K，Beghdadi A，A New Look to Multichannel BlindImage Deconvolution，IEEE Transactions on Image Processing，2009，18（7）：1487~1500
[84] Souidene W，Beghdadi A，Abed-Meraim K，et al，Regularized MRE Methodfor Blind Multichannel Image deconvolution，Record of the Thirty-EighthAsilomar Conference on Signals, Systems and Computers，Pacific Grove, CA,United states，2004，（2）：2233~2237
[85] Chi-Hau Chen，Uvais Qidwai，Ultrasound Attenuation Imaging，Journal ofElectrical and Engineering，2004，55（7-8）：180~187
[86] Goldman L.M，Principles of CT and CT Technology，Journal of NuclearMedicine Technology，2007，35（3）：115~128
[87] Benveniste A，Fabre E，Haar S，Markov Nets: Probabilistic Models forDistributed and Concurrent systems，IEEE Transactions on Automatic Control，2003，48（11）：1936~1950
[88] Cardoso, J.F，Infomax and Maximum Likelihood for Blind Source Separation，IEEE Signal Processing Letters，1997，4（4）：112~114
[89] Comon P，Blind Identification and Source Separation in2×3Under-determinedMixtures，IEEE Transactions on Signal Processing，2004，52（1）：11~22
[90] Amari S，Bornemann J，Vahldieck R，Fast and Accurate Analysis of WaveguideFilters by the Coupled-integral-equations Technique，IEEE Transactions onMicrowave Theory and Techniques，1997，45（9）：1611~1618
[91] Godard D， Self-recovering Equalization and Carrier Tracking inTwo-dimensional Data Communication Systems， IEEE Transactions onCommunications，1980，28（11）：1867~1875
[92] Benveniste A，Goursat M，Ruget G，Robust Identification of a Non-minimumPhase System: Blind Adjustment of a Linear Equalizer in DataCommunications，IEEE Transactions on Automatic Control，1980，25（3）：385~399
[93] Benveniste A，Goursat M，Ruget G，Analysis of Stochastic ApproximationSchemes with Discontinuous and Dependent Forcing Terms with Applications toData Communications Algorithms，IEEE Transactions on Automatic Control，1980，25（6）：1042~1058
[94]杨绿溪，现代数字信号处理，北京：科学出版社，2007，298~317
[95] Hadhoud M.M，Thomas D.W，The Two-Dimensional Adaptive LMS（TDLMS）Algorithm，IEEE Transactions on Circuits and Systems，1988，35（5）：485~494
[96] Pollak I，Willsky A.S，Yan Huang，Nonlinear Evolution Equations as Fast andExact Solvers of Estimation Problems，IEEE Transactions on Signal Processing，2005，53（2）：484~498
[97] Farrell M.D Jr，Mersereau R.M，On the Impact of PCA Dimension Reduction forHyperspectral Detection of Difficult Targets，IEEE Geoscience and RemoteSensing Letters，2005，2（2）：192~195
[98] Yijuan Lu，Qi Tian，Discriminant Subspace Analysis: An Adaptive Approach forImage Classification，IEEE Transactions on Multimedia，2009，11（7）：1289~1300
[99] Dianat R，Kasaei S，Dimension Reduction of Optical Remote Sensing Images viaMinimum Change Rate Deviation Method，IEEE Transactions on Geoscienceand Remote Sensing，2010，48（1）：198~206
[100] Azimi-Sadjadi M.R，Khorasani K，A Model Reduction Method for a Class of2-D Systems，IEEE Transactions on Circuits and Systems-I：FundamentalTheory and Applications，1992，39（1）：28~41
[101]冯燕，何明一，宋江红，基于独立成分分析的高光谱图像数据降维及压缩，电子与信息学报，2007，29（12）：2871~2875
[102] Awate S.P，Whitaker R.T，Unsupervised, Information-Theoretic, AdaptiveImage Filtering for Image Restoration，IEEE Transactions on Pattern Analysisand Machine Intelligence，2006，28（3）：364~376
[103] Suzuki K，Jun Zhang，Jianwu Xu，Massive-Training Artificial Neural NetworkCoupled With Laplacian-Eigenfunction-Based Dimensionality Reduction forComputer-Aided Detection of Polyps in CT Colonography，IEEE Transactionson Medical Imaging，2010，29（11）：1907~1917
[104]黄飞，金伟其，曹峰梅，等，相向运动条件下图像的辐射状退化及其复原研究，电子学报，2005，33（9）：1710~1713
[105]江玲玲，冯象初，殷海青，基于Besov空间的图像盲复原算法，数据采集与处理，2008，23（6）：678~682
[106] Samarasinghe P.D，Kennedy R.A，Minimum Kurtosis CMA Deconvolutionfor Blind Image Restoration．4th International Conference on Information andAutomation for Sustainablity，Zhangjiajie，Hunan，China，2008：271~276
[107] Chan Tony F，Wong C.K，Convergence of the Alternating MinimizationAlgorithm for Blind Deconvolution，Elsevier, Linear and its Application，2000，316（3）：259~285
[108] Vural C, Sethares W.A， Blind Image Deconvolution via DispersionMinimization，Digital Signal Processing，2006，26：137~148
[109] Ye Li，Liu K.J.R，Static and Dynamic Convergence Behavior of Adaptive BlindEqualizers，IEEE Transactions on Signal Processing，1996，44（11）：2736~2745
[110] Mizutani E， Dreyfus SE， Second-order Stagewise Backpropagation forHessian-matrix Analyses and Investi-gation of Negative Curvature，NeuralNetworks，2008，21（2-3）：193~203
[111] Marilli R，Panagiotis T，Enrico Del Re，et al，Constant Modulus BlindEqualization based on Fractional Lower-order Statistics，Signal Processing，2004，84：881~894
[112] Liyi ZHANG, Yunshan SUN, Jingyu ZHANG，et al，Medical CT Image BlindEqualization Algorithm based on Orthogonal Transform， Journal ofComputational Information Systems，2011，7（10）：3455~3461
[113] Glick S.J，Weishi Xia，Iterative Restoration of SPECT Projection Images，IEEETransactions on Nuclear Science，1997，44（2）：204~211
[114] Robini M.C，Magnin I.E，Stochastic Nonlinear Image Restoration using theWavelet Transform，IEEE Transactions on Image Processing，2003，12（8）：890~905
[115] Lun D.P.K，Chan T.C.L，Tai-Chiu Hsung，et al，Efficient Blind ImageRestoration using Discrete Periodic Radon Transform，IEEE Transactions onImage Processing，2004，13（2）：188~200
[116] Junmei Zhong，Huifang Sun，Wavelet-Based Multiscale Anisotropic DiffusionWith Adaptive Statistical Analysis for Image Restoration，IEEE Transactions onCircuits and Systems-I: Regular Papers，2008，55（9）：2716~2725
[117] Bin Yang，Shutao Li，Multifocus Image Fusion and Restoration with SparseRepresentation，IEEE Transactions on Instrumentation and Measurement，2010，59（4）：884~892
[118] Nikolova M， Ng M.K， Chi-Pan Tam， Fast Nonconvex NonsmoothMinimization Methods for Image Restoration and Reconstruction， IEEETransactions on Image Processing，2010，19（12）：3073~3088
[119] Chiang H.H，Nikias C.L，Adaptive Deconvolution and Identification ofNonminimum Phase FIR Systems based on Cumulants，IEEE Transactions onAutomatic Control，1990，35（1）：36~47
[120] Osareh A，Shadgar B，Markham R，A Computational-Intelligence-BasedApproach for Detection of Exudates in Diabetic Retinopathy Images，IEEETransactions on Information Technology in Biomedicine，2009，13（4）：535~545
[121] Tristan-Vega A，Arribas J.I，A Radius and Ulna TW3Bone Age AssessmentSystem，IEEE Transactions on Biomedical Engineering，2008，55（5）：1463~1476
[122] Tao Song，Jamshidi M.M，Lee R.R，et al，A Modified Probabilistic NeuralNetwork for Partial Volume Segmentation in Brain MR Image， IEEETransactions on Neural Networks，2007，18（5）：1424~1432
[123] Chuan-Yu Chang， Yue-Fong Lei， Chin-Hsiao Tseng， et al， ThyroidSegmentation and Volume Estimation in Ultrasound Images，IEEE Transactionson Biomedical Engineering，2010，57（6）：1348~1357
[124] Sheng-Fang Huang，Chang Ruey-Feng，Woo Kyung Moon，et al，Analysis ofTumor Vascularity using Three-Dimensional Power Doppler UltrasoundImages，IEEE Transactions on Medical Imaging，2008，27（3）：320~330
[125] Shen-Shu Xiong，Zhao-Ying Zhou，Neural Filtering of Colored Noise based onKalman Filter Structure， IEEE Transactions on Instrumentation andMeasurement，2003，52（3）：742~747
[126] Zhou Y.T，Chellappa R，Vaid A，et al，Image Restoration using a NeuralNetwork，IEEE Transactions on Acoustics, Speech and Signal Processing，1988，36（7）：1141~1151
[127] Paik J.K，Katsaggelos A.K，Image Restoration using a Modified HopfieldNetwork，IEEE Transactions on Image Processing，1992，1（1）：49~63
[128] Yi Sun，Hopfield Neural Network based Algorithms for Image Restoration andReconstruction. I. Algorithms and Simulations，IEEE Transactions on SignalProcessing，2000，48（7）：2105~2118
[129] Yi Sun，Hopfield Neural Network based Algorithms for Image Restoration andReconstruction. II. Performance Analysis， IEEE Transactions on SignalProcessing，2000，48（7）：2119~2131
[130] Liu H.J，Sun Y，Blind Bilevel Image Restoration using Hopfield NeuralNetworks，IEEE International Conference on Neural Networks，1993，3：1656-1661
[131] Cheema T.A，Qureshi I.M，Jalil, A，et al，Space-variant Neural NetworkApproach to Blind Image Deconvolution，2006IEEE International Conferenceon Multitopic，Islamabad，2005：120~127
[132] Aizenberg I，Paliy D.V，Zurada J.M，et al，Blur Identification by MultilayerNeural Network Based on Multivalued Neurons，IEEE Transactions on NeuralNetworks，2008，19（5）：883~898
[133] Liwei Zhang，Yaping Zhang，A New Image Restoration Algorithm based onWindow Roaming and RBF Neural Network，2011International Conference onMechatronics and Automation，Beijing，China，2011：2009~2013
[134] Wen-Hao Lee，Shang-Hong Lai，Chia-Lun Chen，Iterative Blind Image MotionDeblurring via Learning a No-Reference Image Quality Measure， IEEEInternational Conference on Image Processing，San Antonio, TX，2007，4：405~408
[135] Zhang Q，Benveniste A，Wavelet Networks，IEEE Transactions on NeuralNetworks，1992，3（6）：889~898
[136] Lo S.C.B，Huai Li，Freedman M.T，Optimization of Wavelet Decompositionfor Image Compression and Feature Preservation，IEEE Transactions on MedicalImaging，2003，22（9）：1141-1151
[137] Soo-Chang Pei，Chien-Cheng Tseng，Ching-Yung Lin，Wavelet Transform andScale Space Filtering of Fractal Images， IEEE Transactions on ImageProcessing，1995，4（5）：682-687
[138] Wei Qian，Clarke L.P，Wavelet-Based Neural Network with Fuzzy-logicAdaptivity for Nuclear Image Restoration，Proceedings of the IEEE，1996，84（10）：1458~1473
[139] Zineddin B，Zidong Wang，Xiaohui Liu，Cellular Neural Networks, theNavier–Stokes Equation, and Microarray Image Reconstruction， IEEETransactions on Image Processing，2011，20（11）：3296~3301
[140] Xuena Jiang，Yangyang Liu，Shoujue Wang，A Novel Geometric Algorithm forBlind Image Restoration Based on High-Dimensional Space，2nd InternationalCongress on Mage and Signal Processing， Tianjin，China，2009：1~5
[141] Mukherjee S.S，Chowdhury R，Bhattacharyya S，Image Restoration using aMultilayered Quantum Backpropagation Neural Network，2011InternationalConference on Computational Intelligence and Communication Networks，Gwalior，India，2011，426~430
[142]刘炜，孙丰荣，梅良模，基于信号多尺度边缘表示的CT医学图像增强，中国图象图形学报，2005，1（2）：207~212
[143] Li Ping，Xiaoling Huang，Nam Phamdo，Zigzag Codes and ConcatenatedZigzag Codes，IEEE Transactions on Information Theory，2001,47（2）：800~807
[144] Wei Wu，Guorui Feng，Zhengxue Li，et al，Deterministic Convergence of anOnline Gradient Method for Bp Neural Networks，IEEE Transactions onNeural Networks，2005，16：533~540
[145] Wei Wu，Hongmei Shao，Di Qu，Strong Convergence for Gradient Methodsfor BP Network Training， Processing International Conference on NeuralNetworks and Brains，2005，1：332~334
[146] Chan W.L，So A.T.P，Lai L.L，Initial Applications of Complex Artificial NeuralNetworks to Load-flow Analysis，IEE Proceedings Generation, Transmissionand Distribution，2000，147（6）：361~366
[147] Cheolwoo You，Daesik Hong，Adaptive Equalization using the ComplexBackpropagation Algorithm， IEEE International Conference on NeuralNetworks Washington, DC, USA，1996，（4）：2136~2141
[148] Cheolwoo You，Daesik Hong，Nonlinear Blind Equalization Schemes usingComplex-valued Multilayer Feed-forward Neural Networks，IEEE Transactionson Signal Processing，1998，9（6）：1442~1455
[02]刘涛，郭军，盲图像恢复中的二维盲均衡研究，电子与信息学报，2006，28（6）：1013~1015
[03]张金玉，黄先祥，谢伟达，基于盲均衡理论的弱冲击故障的检测研究，机械科学与技术，2008，27（8）：996~999
[04] Xiaohua Li，Blind Channel Estimation and Equalization in Wireless SensorNetworks Based on Correlations among Sensors，IEEE Transactions on SignalProcessing，2005，53（4）：1511~1519
[05] Bruneau M，Reinhorn A.M， Constantinou M.C， et al，The University atBuffalo（UB）Node of the NEES Network–a Versatile High Performance TestingFacility towards Real-time Dynamic Hybrid Testing，12th European Conferenceon Earthquake Engineering London，UK，2002：1~8
[06] Pavlovic G， Tekalp A.M， Maximum Likelihood Parametric based on aContinuous Spatial Blur Identification Domain Model，IEEE Transactions onImage Processing，1992，I（4）：496~504
[07] Li D，Mersereau R.M，Simske S，Blind Image Deconvolution Through SupportVector Regression，IEEE Transactions on Neural Networks，2007，18（3）：931~935
[08] Jingyan Xu，Taguchi K，Tsui B.M.W，Statistical Projection Completion in X-rayCT Using Consistency Conditions，IEEE Transactions on Medical Imaging，2010，29（8）：1528~1540
[09] Figueiredo M.A.T，Bioucas-Dias J.M，Restoration of Poissonian Images usingAlternating Direction Optimization，IEEE Transactions on Image Processing，2010，19（12）：3133~3145
[10] Reeves S.J，Mersereau R.M，Blur Identification by the Method of GeneralizedCross-validation，IEEE Transaction on Image Processing，1992，1（3）：301~311
[11] Nguyen N，Golub G，Milanfar P，Blind restoration/superresolution withgeneralized cross-validation using Gauss-type quadrature rules，Record of theThirty-Third Asilomar Conference on Signals, Systems, and Computers，PacificGrove, CA, USA，1999，（2）：1257~1261
[12] Haiyong Liao， Ng M.K， Blind Deconvolution Using GeneralizedCross-Validation Approach to Regularization Parameter Estimation， IEEETransactions on Image Processing，2011，20（3）：670~680
[13] Tsumuraya F，Miura N，Baba N，Iterative Blind Deconvolution Method usingLucy’s Algorithm，Astronomy and Astrophysics，1994，282：669~708
[14] Likas A.C，Galatsanos N.P，A Variational Approach for Bayesian Blind ImageDeconvolution，IEEE Transactions on Signal Processing，2004，52（8）：2222~2233
[15] Gupta S，Chauhan R.C，Saxena S.C，Locally Adaptive Wavelet DomainBayesian Processor for Denoising Medical Ultrasound Images using SpeckleModelling based on Rayleigh Distribution，IEE Proceeding of Vision, Image andSignal Processing，2005，152（1）：129~135
[16] Tzikas D.G， Likas A.C， Galatsanos N.P， Variational Bayesian SparseKernel-Based Blind Image Deconvolution With Student’s-t Priors， IEEETransactions on Image Processing，2009，18（4）：753~764
[17] Bishop T.E，Molina R，Hopgood J.R，Blind Restoration of Blurred Photographsvia AR Modelling and MCMC，15th IEEE International Conference on ImageProcessing，San Diego, CA, United states，2008：669~672
[18] Min Dai，Cheng Peng，Chan A.K，et al，Bayesian Wavelet Shrinkage with EdgeDetection for SAR Image Despeckling，IEEE Transactions on Geoscience andRemote Sensing，2004，42（8）：1642~1648
[19] Babacan S.D，Jingnan Wang，Molina R，et al，Bayesian Blind DeconvolutionFrom Differently Exposed Image Pairs， IEEE Transactions on ImageProcessing，2010，19（11）：2874~2888
[20] Karayiannis N.B，Venetsanopoulos A.N，Regularization Theory in ImageRestoration-The Stabilizing Functional Approach， IEEE Transactions onAcoustics, Speech, and Signal Processing，1990，38（7）：1155~1179
[21] Kim-Hui Yap，Ling Guan，A Computational Reinforced Learning Scheme toBlind Image Deconvolution，IEEE Transactions on Evolutionary Computation，2002，6（1）：2~15
[22] Michailovich O.V，Adam D，A Novel Approach to the2-D Blind DeconvolutionProblem in Medical Ultrasound，IEEE Transactions on Medical Imaging，2005，24（1）：86~104
[23] Belekos S.P，Galatsanos N.P，Katsaggelos A.K，Maximum a Posteriori VideoSuper-Resolution Using a New Multichannel Image Prior，IEEE Transactions onImage Processing，2010，19（6）：1451~1464
[24] Humphrey D，Taubman D，A Filtering Approach to Edge Preserving MAPEstimation of Images，IEEE Transactions on Image Processing，2011，20（5）：1234~1248
[25] Ming Jiang，Ge Wang，Skinner，M.W，et al，Blind Deblurring of Spiral CTImages，IEEE Transactions on Medical Imaging，2003，22（7）：837~845
[26] Li Chen，Kim-Hui Yap，A Soft Double Regularization Approach to ParametricBlind Image Deconvolution，IEEE Transactions on Image Processing，2005，14（5）：624~633
[27] Li Chen，Kim-Hui Yap，Efficient Discrete Spatial Techniques for Blur SupportIdentification in Blind Image Deconvolution，IEEE Transactions on SignalProcessing，2006，54（4）：1557~1562
[28] Sanches J.M，Nascimento J.C，Marques J.S，Medical Image Noise Reductionusing the Sylvester–Lyapunov Equation， IEEE Transactions on ImageProcessing，2008，17（9）：1522~1539
[29] Ayers G.R， Dainty J.C， Iterative Blind Deconvolution Method and ItsApplications，Optics Letters，1988，13（7）：547~549
[30] Lam E.Y， Blind Bi-Level Image Restoration with Iterated QuadraticProgramming，IEEE Transactions on Circuits and Systems-II: Express Briefs，2007，54（1）：52~56
[31] Szolgay D，Szirányi T，Optimal Stopping Condition for Iterative ImageDeconvolution by New Orthogonality Criterion，Electronics Letters，2011，47（7）：442~444
[32] Kundur D，Hatzinakos D，A Novel Blind Deconvolution Scheme for BlindImage Restoration using Recursive Filtering，IEEE Transactions on ImageProcessing，1998，46（2）：375~390
[33]薛梅，杨绿溪，邹采荣，用于含噪二值图像的改进NAS-RIF图像盲复原算法，数据采集与处理，2002，17（2）：156~160
[34] Hongzhi Wang，Sun Qi，Yuanyuan Liu，An Improved NAS-RIF Algorithmbased on the Lifting Scheme for Blind Image Restoration， Pacific-AsiaConference on Circuits, Communications and Systems，Chengdu，China，2009：423~426
[35] Chunpeng Song，Hongzhi Wang，Liliang He，et al，An Improved NAS-RIF BlindImage Restoration based on Higher Order Statistics，20068th InternationalConference on Signal Processing，Guilin，China，2006，（4）：553~556
[36] Kundur D，Hatzinakos D，Leung H，Robust Classification of Blurred Imagery，IEEE Transactions on Image Processing，2000，9（2）：243~255
[37] Kundur D，Hatzinakos D，On the Use of Lyapunov Criteria to Analyze theConvergence of Blind Deconvolution Algorithms，IEEE Transactions on SignalProcessing，1998，46（11）：2918~2925
[38] Chin Ann Ong，Chambers J.A，An Enhanced NAS-RIF Algorithm for BlindImage Deconvolution，IEEE Transactions on image processing，1999，8（7）：988~992
[39] Matsuyama M，Tanji Y，Tanaka M，Enhancing the Ability of NAS-RIFAlgorithm for Blind Image Deconvolution，ISCAS2000IEEE InternationalSymposium on Circuit and Systems，Geneva, Switzerland，2000，（4）：553~556
[40] McCallum B.C， Blind Deconvolution by Simulated Annealing， OpticsCommunications，1990，75（2）：101~105
[41] Leipo Yan，Lipo Wang，Image Restoration using Chaotic Simulated Annealing，Proceedings of the International Joint Conference on Neural Networks，Portland,OR, United states，2003，4：3060-3064
[42]张红英，彭启琮，一种改进的模拟退火图像盲复原算法，电子科技大学学报，2008，35（5）：767~769，787
[43] Vogel C.R，Oman M.E，Fast Robust Total Variation Based Reconstruction ofNoisy Blurred Images，IEEE Transactions on Image Processing，1998，7（6）：813~824
[44] Chan T.F，Chiu-Kwong Wong，Total Variation Blind Image Deconvolution，IEEE Transactions on Image Processing，1998，7（3）：370~375
[45] Hongwei Zheng，Hellwich O，An Edge-Driven Total Variation Approach toImage Deblurring and Denoising， Proceedings of the First InternationalConference on Innovative Computing, Information and Control，Beijing，China，2006，（2）：705~710
[46]张航，罗大庸，一种改进的全变差盲图像复原方法，电子学报，2005，33（7）：1288~1290
[47] Chan R.H，Yiqiu Dong，Hintermuller M，An Efficient Two-Phase L1-TV Methodfor Restoring Blurred Images with Impulse Noise，IEEE Transactions on ImageProcessing，2011，19（7）：1731~1739
[48] Michailovich O.V，An Iterative Shrinkage Approach to Total-Variation ImageRestoration，IEEE Transactions on Image Processing，2011，20（5）：1281~1299
[49] Wang S，Liu Z.W，Dong W.S，et al，Total Variation based Image Deblurringwith Nonlocal Self-similarity Constraint，Electronics Letters，2011，47（16）：916~918
[50] Kim-Hui Yap，Ling Guan，Wanquan Liu，A Recursive Soft-Decision Approachto Blind Image Deconvolution，IEEE Transactions on Signal Processing，2003，51（2）：515~526
[51] Cheema T.A，Qureshi I.M，Hussain A，Blind Image Deconvolution usingSpace-variant Neural Network Approach，Electronics Letters，2005，41（6）：308~309
[52] Russo F，Noise Removal from Image Data Using Recursive Neurofuzzy Filters，IEEE Transactions on Instrumentation and Measurement，2000，49（2）：307~314
[53] Senel H.G，Peters R.A II，Dawant B，Topological Median Filters，IEEETransactions on Image Processing，2002，11（2）：89~104
[54] Van De Ville D，Nachtegael M，Van der Weken D，et al，Noise Reduction byFuzzy Image Filtering，IEEE Transactions on Fuzzy Systems，2003，11（4）：429~436
[55] Haixiang Xu，Guangxi Zhu，Fuyuan Peng，et al，Adaptive Fuzzy Switching Filterfor Images Corrupted by Impulse Noise，2004International Conference onCommunications, Circuits and Systems，2004，（2）：792~795
[56]屈志毅，沃炎，任志宏，基于交替迭代和神经网络的盲目图像恢复，计算机学报，2000，23（4）：410~413
[57] Adam D，Michailovich O，Blind Deconvolution of Ultrasound Sequences UsingNonparametric Local Polynomial Estimates of the Pulse，IEEE Transactions onBiomedical Engineering，2002，49（2）：118~131
[58] Wan S，Raju, B.I，Srinivasan M.A，Robust Deconvolution of High-FrequencyUltrasound Images Using Higher-Order Spectral Analysis and Wavelets，IEEETransactions on Ultrasonics, Ferroelectrics, and Frequency Control，2003，50（10）：1286~1295
[59] Yew Kun Chee，Yeng Chai Soh，A Robust Kalman Filter Design for ImageRestoration，Proceedings of Acoustics, Speech, and Signal Processing，SaltLake，UT，United states，2001，（3）：1825~1828
[60] Jing Wang，Ge Wang，Ming Jiang，Blind deblurring of Spiral CT Images basedon ENR and Wiener filter，Journal of X-Ray Science and Technology，2005，13：49~60
[61] Lane R.G，Bates R.H.T，Automatic Multidimensional Deconvolution，Journal of Optical Society of America，1987，4（1）：180~188
[62] Ghigila D.C， Romero L.A， Mastin G.A， Systematic Approach toTwo-dimensional Blind Image Deconvolution by Zero-sheets Separation，Journal of Optical Society of America，1993，10（5）：1024~1036
[63] Nakagaki R， Katsaggelos A.K， A VQ-Based Blind Image RestorationAlgorithm，IEEE Transactions on Image Processing，2003，12（9）：1044~1053
[64] Tsung-Ying Sun，Chan-Cheng Liu，Yu-Peng Jheng，et al，Blind ImageDeconvolution via Particle Swarm Optimization with Entropy Evaluation，Eighth International Conference on Intelligent Systems Design andApplications，Kaohsiung，Taiwan，2008：265~270
[65] Johnson Eric G，Abusshagur Mustafa A G，Image Deconvolution using a MicroGenetic Algorithm，Optics Communications，1997，140（1-3）：6~10
[66] Yang-Chih Lai，Chih-Li Huo，Yu-Hsiang Yu，et al，PSO-based Estimation forGaussian Blur in Blind Image Deconvolution Problem，2011IEEE InternationalConference on Fuzzy Systems，Taipei，Taiwan，2011：1143~1148
[67]张煜东，吴乐南，改进免疫算法用于图像复原，光学精密工程，2009，17（2）：417~425
[68] Umeyama S， Blind Deconvolution of Images using Gabor Filters andIndependent Component Analysis，4th International Symposium on IndependentComponent Analysis and Blind Signal Separation，Nara，Japan，2003：319~324
[69] Dylov D. V，Waller L，Fleischer J. W，Nonlinear Restoration of Diffused Imagesvia Seeded Instability，IEEE Journal of Selected Topics in Quantum Electronics，2012，18（2）：916~925
[70] Gazzah H，Regalia P.A，Delmas J.P，et al，A Blind Multichannel IdentificationAlgorithm Robust to Order Overestimation，IEEE Transactions on SignalProcessing，2002，50（6）：1449~1458
[71] Hendekli C.M， Ertuzun, A， Design of a Multichannel Two-dimensionalDelta-domain Lattice Filter for Noise Removal，IEEE Transactions on SignalProcessing，2001，49（7）：1581~1593
[72] Vrhel M.J， Unser M， Multichannel Restoration with Limited a PrioriInformation，IEEE Transactions on Image Processing，1999，8（4）：527~536
[73] Elad M，Feuer A，Restoration of a Single Superresolution Image from SeveralBlurred, Noisy and Undersampled Measured Images，IEEE Transactions onImage Processing，1997，6（12）：1646~1658
[74] Miura N，Baba N，Extended-object Reconstruction with Sequential use of theIterative Blind Deconvolution Method，Optics Communications，1992，89（5-6）：375~379
[75] Schulz T.J，Multiframe Blind Deconvolution of Astronomical Images，Journal ofthe Optical Society of America A: Optics and Image Science and Vision，1993，10（5）：1064~1073
[76] Moon Gi Kang，Katsaggelos A.K，Simultaneous Multichannel Image Restorationand Estimation of the Regularization Parameters，IEEE Transactions on ImageProcessing，1997，6（5）：774~778
[77] Hung-Ta Pai，Bovik A.C，On Eigenstructure-based Direct Multichannel BlindImage Restoration，IEEE Transactions on Image Processing，2001，10（10）：1434~1446
[78] Sroubek F，Flusser J，Multichannel Blind Iterative Image Restoration，IEEETransactions on Image Processing，2003，12（9）：1094~1106
[79] Giannakis G.B，Heath R.W Jr，Blind Identification of Multichannel FIR Blursand Perfect Image Restoration，IEEE Transactions on Image Processing，2000，9（11）：1877~1896
[80] Hung-Ta Pai，Bovik A.C，Exact Multichannel Blind Image Restoration，IEEESignal Processing Letters，1997，4（8）：217~220
[81] Chow T.W.S，Xiao-Dong Li，Kai-Tat Ng，Double-regularization Approach forBlind Restoration of Multichannel Imagery，IEEE Transactions on Circuits andSystems I: Fundamental Theory and Applications，2001，48（9）：1075~1085
[82] Panci G， Campisi P， Colonnese S， et al， Multichannel Blind ImageDeconvolution Using the Bussgang Algorithm: Spatial and MultiresolutionApproaches，IEEE Transactions on Image Processing，2003，1（211）：1324~1337
[83] Souidene W，Abed-Meraim K，Beghdadi A，A New Look to Multichannel BlindImage Deconvolution，IEEE Transactions on Image Processing，2009，18（7）：1487~1500
[84] Souidene W，Beghdadi A，Abed-Meraim K，et al，Regularized MRE Methodfor Blind Multichannel Image deconvolution，Record of the Thirty-EighthAsilomar Conference on Signals, Systems and Computers，Pacific Grove, CA,United states，2004，（2）：2233~2237
[85] Chi-Hau Chen，Uvais Qidwai，Ultrasound Attenuation Imaging，Journal ofElectrical and Engineering，2004，55（7-8）：180~187
[86] Goldman L.M，Principles of CT and CT Technology，Journal of NuclearMedicine Technology，2007，35（3）：115~128
[87] Benveniste A，Fabre E，Haar S，Markov Nets: Probabilistic Models forDistributed and Concurrent systems，IEEE Transactions on Automatic Control，2003，48（11）：1936~1950
[88] Cardoso, J.F，Infomax and Maximum Likelihood for Blind Source Separation，IEEE Signal Processing Letters，1997，4（4）：112~114
[89] Comon P，Blind Identification and Source Separation in2×3Under-determinedMixtures，IEEE Transactions on Signal Processing，2004，52（1）：11~22
[90] Amari S，Bornemann J，Vahldieck R，Fast and Accurate Analysis of WaveguideFilters by the Coupled-integral-equations Technique，IEEE Transactions onMicrowave Theory and Techniques，1997，45（9）：1611~1618
[91] Godard D， Self-recovering Equalization and Carrier Tracking inTwo-dimensional Data Communication Systems， IEEE Transactions onCommunications，1980，28（11）：1867~1875
[92] Benveniste A，Goursat M，Ruget G，Robust Identification of a Non-minimumPhase System: Blind Adjustment of a Linear Equalizer in DataCommunications，IEEE Transactions on Automatic Control，1980，25（3）：385~399
[93] Benveniste A，Goursat M，Ruget G，Analysis of Stochastic ApproximationSchemes with Discontinuous and Dependent Forcing Terms with Applications toData Communications Algorithms，IEEE Transactions on Automatic Control，1980，25（6）：1042~1058
[94]杨绿溪，现代数字信号处理，北京：科学出版社，2007，298~317
[95] Hadhoud M.M，Thomas D.W，The Two-Dimensional Adaptive LMS（TDLMS）Algorithm，IEEE Transactions on Circuits and Systems，1988，35（5）：485~494
[96] Pollak I，Willsky A.S，Yan Huang，Nonlinear Evolution Equations as Fast andExact Solvers of Estimation Problems，IEEE Transactions on Signal Processing，2005，53（2）：484~498
[97] Farrell M.D Jr，Mersereau R.M，On the Impact of PCA Dimension Reduction forHyperspectral Detection of Difficult Targets，IEEE Geoscience and RemoteSensing Letters，2005，2（2）：192~195
[98] Yijuan Lu，Qi Tian，Discriminant Subspace Analysis: An Adaptive Approach forImage Classification，IEEE Transactions on Multimedia，2009，11（7）：1289~1300
[99] Dianat R，Kasaei S，Dimension Reduction of Optical Remote Sensing Images viaMinimum Change Rate Deviation Method，IEEE Transactions on Geoscienceand Remote Sensing，2010，48（1）：198~206
[100] Azimi-Sadjadi M.R，Khorasani K，A Model Reduction Method for a Class of2-D Systems，IEEE Transactions on Circuits and Systems-I：FundamentalTheory and Applications，1992，39（1）：28~41
[101]冯燕，何明一，宋江红，基于独立成分分析的高光谱图像数据降维及压缩，电子与信息学报，2007，29（12）：2871~2875
[102] Awate S.P，Whitaker R.T，Unsupervised, Information-Theoretic, AdaptiveImage Filtering for Image Restoration，IEEE Transactions on Pattern Analysisand Machine Intelligence，2006，28（3）：364~376
[103] Suzuki K，Jun Zhang，Jianwu Xu，Massive-Training Artificial Neural NetworkCoupled With Laplacian-Eigenfunction-Based Dimensionality Reduction forComputer-Aided Detection of Polyps in CT Colonography，IEEE Transactionson Medical Imaging，2010，29（11）：1907~1917
[104]黄飞，金伟其，曹峰梅，等，相向运动条件下图像的辐射状退化及其复原研究，电子学报，2005，33（9）：1710~1713
[105]江玲玲，冯象初，殷海青，基于Besov空间的图像盲复原算法，数据采集与处理，2008，23（6）：678~682
[106] Samarasinghe P.D，Kennedy R.A，Minimum Kurtosis CMA Deconvolutionfor Blind Image Restoration．4th International Conference on Information andAutomation for Sustainablity，Zhangjiajie，Hunan，China，2008：271~276
[107] Chan Tony F，Wong C.K，Convergence of the Alternating MinimizationAlgorithm for Blind Deconvolution，Elsevier, Linear and its Application，2000，316（3）：259~285
[108] Vural C, Sethares W.A， Blind Image Deconvolution via DispersionMinimization，Digital Signal Processing，2006，26：137~148
[109] Ye Li，Liu K.J.R，Static and Dynamic Convergence Behavior of Adaptive BlindEqualizers，IEEE Transactions on Signal Processing，1996，44（11）：2736~2745
[110] Mizutani E， Dreyfus SE， Second-order Stagewise Backpropagation forHessian-matrix Analyses and Investi-gation of Negative Curvature，NeuralNetworks，2008，21（2-3）：193~203
[111] Marilli R，Panagiotis T，Enrico Del Re，et al，Constant Modulus BlindEqualization based on Fractional Lower-order Statistics，Signal Processing，2004，84：881~894
[112] Liyi ZHANG, Yunshan SUN, Jingyu ZHANG，et al，Medical CT Image BlindEqualization Algorithm based on Orthogonal Transform， Journal ofComputational Information Systems，2011，7（10）：3455~3461
[113] Glick S.J，Weishi Xia，Iterative Restoration of SPECT Projection Images，IEEETransactions on Nuclear Science，1997，44（2）：204~211
[114] Robini M.C，Magnin I.E，Stochastic Nonlinear Image Restoration using theWavelet Transform，IEEE Transactions on Image Processing，2003，12（8）：890~905
[115] Lun D.P.K，Chan T.C.L，Tai-Chiu Hsung，et al，Efficient Blind ImageRestoration using Discrete Periodic Radon Transform，IEEE Transactions onImage Processing，2004，13（2）：188~200
[116] Junmei Zhong，Huifang Sun，Wavelet-Based Multiscale Anisotropic DiffusionWith Adaptive Statistical Analysis for Image Restoration，IEEE Transactions onCircuits and Systems-I: Regular Papers，2008，55（9）：2716~2725
[117] Bin Yang，Shutao Li，Multifocus Image Fusion and Restoration with SparseRepresentation，IEEE Transactions on Instrumentation and Measurement，2010，59（4）：884~892
[118] Nikolova M， Ng M.K， Chi-Pan Tam， Fast Nonconvex NonsmoothMinimization Methods for Image Restoration and Reconstruction， IEEETransactions on Image Processing，2010，19（12）：3073~3088
[119] Chiang H.H，Nikias C.L，Adaptive Deconvolution and Identification ofNonminimum Phase FIR Systems based on Cumulants，IEEE Transactions onAutomatic Control，1990，35（1）：36~47
[120] Osareh A，Shadgar B，Markham R，A Computational-Intelligence-BasedApproach for Detection of Exudates in Diabetic Retinopathy Images，IEEETransactions on Information Technology in Biomedicine，2009，13（4）：535~545
[121] Tristan-Vega A，Arribas J.I，A Radius and Ulna TW3Bone Age AssessmentSystem，IEEE Transactions on Biomedical Engineering，2008，55（5）：1463~1476
[122] Tao Song，Jamshidi M.M，Lee R.R，et al，A Modified Probabilistic NeuralNetwork for Partial Volume Segmentation in Brain MR Image， IEEETransactions on Neural Networks，2007，18（5）：1424~1432
[123] Chuan-Yu Chang， Yue-Fong Lei， Chin-Hsiao Tseng， et al， ThyroidSegmentation and Volume Estimation in Ultrasound Images，IEEE Transactionson Biomedical Engineering，2010，57（6）：1348~1357
[124] Sheng-Fang Huang，Chang Ruey-Feng，Woo Kyung Moon，et al，Analysis ofTumor Vascularity using Three-Dimensional Power Doppler UltrasoundImages，IEEE Transactions on Medical Imaging，2008，27（3）：320~330
[125] Shen-Shu Xiong，Zhao-Ying Zhou，Neural Filtering of Colored Noise based onKalman Filter Structure， IEEE Transactions on Instrumentation andMeasurement，2003，52（3）：742~747
[126] Zhou Y.T，Chellappa R，Vaid A，et al，Image Restoration using a NeuralNetwork，IEEE Transactions on Acoustics, Speech and Signal Processing，1988，36（7）：1141~1151
[127] Paik J.K，Katsaggelos A.K，Image Restoration using a Modified HopfieldNetwork，IEEE Transactions on Image Processing，1992，1（1）：49~63
[128] Yi Sun，Hopfield Neural Network based Algorithms for Image Restoration andReconstruction. I. Algorithms and Simulations，IEEE Transactions on SignalProcessing，2000，48（7）：2105~2118
[129] Yi Sun，Hopfield Neural Network based Algorithms for Image Restoration andReconstruction. II. Performance Analysis， IEEE Transactions on SignalProcessing，2000，48（7）：2119~2131
[130] Liu H.J，Sun Y，Blind Bilevel Image Restoration using Hopfield NeuralNetworks，IEEE International Conference on Neural Networks，1993，3：1656-1661
[131] Cheema T.A，Qureshi I.M，Jalil, A，et al，Space-variant Neural NetworkApproach to Blind Image Deconvolution，2006IEEE International Conferenceon Multitopic，Islamabad，2005：120~127
[132] Aizenberg I，Paliy D.V，Zurada J.M，et al，Blur Identification by MultilayerNeural Network Based on Multivalued Neurons，IEEE Transactions on NeuralNetworks，2008，19（5）：883~898
[133] Liwei Zhang，Yaping Zhang，A New Image Restoration Algorithm based onWindow Roaming and RBF Neural Network，2011International Conference onMechatronics and Automation，Beijing，China，2011：2009~2013
[134] Wen-Hao Lee，Shang-Hong Lai，Chia-Lun Chen，Iterative Blind Image MotionDeblurring via Learning a No-Reference Image Quality Measure， IEEEInternational Conference on Image Processing，San Antonio, TX，2007，4：405~408
[135] Zhang Q，Benveniste A，Wavelet Networks，IEEE Transactions on NeuralNetworks，1992，3（6）：889~898
[136] Lo S.C.B，Huai Li，Freedman M.T，Optimization of Wavelet Decompositionfor Image Compression and Feature Preservation，IEEE Transactions on MedicalImaging，2003，22（9）：1141-1151
[137] Soo-Chang Pei，Chien-Cheng Tseng，Ching-Yung Lin，Wavelet Transform andScale Space Filtering of Fractal Images， IEEE Transactions on ImageProcessing，1995，4（5）：682-687
[138] Wei Qian，Clarke L.P，Wavelet-Based Neural Network with Fuzzy-logicAdaptivity for Nuclear Image Restoration，Proceedings of the IEEE，1996，84（10）：1458~1473
[139] Zineddin B，Zidong Wang，Xiaohui Liu，Cellular Neural Networks, theNavier–Stokes Equation, and Microarray Image Reconstruction， IEEETransactions on Image Processing，2011，20（11）：3296~3301
[140] Xuena Jiang，Yangyang Liu，Shoujue Wang，A Novel Geometric Algorithm forBlind Image Restoration Based on High-Dimensional Space，2nd InternationalCongress on Mage and Signal Processing， Tianjin，China，2009：1~5
[141] Mukherjee S.S，Chowdhury R，Bhattacharyya S，Image Restoration using aMultilayered Quantum Backpropagation Neural Network，2011InternationalConference on Computational Intelligence and Communication Networks，Gwalior，India，2011，426~430
[142]刘炜，孙丰荣，梅良模，基于信号多尺度边缘表示的CT医学图像增强，中国图象图形学报，2005，1（2）：207~212
[143] Li Ping，Xiaoling Huang，Nam Phamdo，Zigzag Codes and ConcatenatedZigzag Codes，IEEE Transactions on Information Theory，2001,47（2）：800~807
[144] Wei Wu，Guorui Feng，Zhengxue Li，et al，Deterministic Convergence of anOnline Gradient Method for Bp Neural Networks，IEEE Transactions onNeural Networks，2005，16：533~540
[145] Wei Wu，Hongmei Shao，Di Qu，Strong Convergence for Gradient Methodsfor BP Network Training， Processing International Conference on NeuralNetworks and Brains，2005，1：332~334
[146] Chan W.L，So A.T.P，Lai L.L，Initial Applications of Complex Artificial NeuralNetworks to Load-flow Analysis，IEE Proceedings Generation, Transmissionand Distribution，2000，147（6）：361~366
[147] Cheolwoo You，Daesik Hong，Adaptive Equalization using the ComplexBackpropagation Algorithm， IEEE International Conference on NeuralNetworks Washington, DC, USA，1996，（4）：2136~2141
[148] Cheolwoo You，Daesik Hong，Nonlinear Blind Equalization Schemes usingComplex-valued Multilayer Feed-forward Neural Networks，IEEE Transactionson Signal Processing，1998，9（6）：1442~1455