GABOR小波神经网络算法及其在灰度图象目标识别中的应用研究
|
摘要
本文主要研究的是Gabor小波神经网络算法及其在灰度图像目标识别中的应用。研究涉及神经网络理论、小波分析理论和小波神经网络理论及其它们在目标识别中的应用技术。论文以基于多CPU神经网络目标识别系统的设计与实现为宗旨，在理论和实践两个方面研究和探讨了系统实现过程中所涉及的关键算法和关键技术。论文包括以下诸方面内容：
对目前国际、国内BP神经网络算法、小波变换理论、Gabor小波神经网络算法以及在目标识别中的应用现状进行了综述，对存在的问题进行了剖析。同时以灰度图像目标识别为主线，阐述与论文直接相关的基本理论和关键技术。
详细分析了多层前馈型神经网络描述及训练算法机理，从数学的角度推导了误差逆传播算法(BP算法)，同时指出了BP算法存在的问题。构建了一种用于多目标识别的改进的BP算法。
阐述了Gabor滤波器原理。给出了Gabor小波滤波器的表达式。根据理论分析和实际需要，设计了多通道Gabor滤波器，提取了灰度图像目标纹理特征。相应地给出了一个基于多通道Gabor滤波器特征神经网络识别算法。
给出了相位叠加不变量pc的定义。从理论上详细探讨了用Gabor小波求解相位叠加不变量PC的数学过程，给出了相应的PC不变量的公式。
详细分析了log gabor函数性能，给出了用log gabor求相位叠加(PC)不变量的修正公式。最后，设计了基于相位叠加(PC)不变量的神经网络目标识别新算法。
阐述了三种不同的目标表示法，即基于特征的目标描述法、基于模板的目标描述法和基于Gabor小波神经网络目标描述法，同时分析了三种表示法的优缺点。详细阐述了Gabor小波神经网络构建方法，重点构建了Gabor小波神经网络的训练算法，给出了理论分析和算法的具体步骤。
本文的主要研究思想就是：运用并行的Gabor小波神经网络算法对灰度图像目标进行实时识别。研究的思路：根据前馈神经网络模型(BP网络)和Gabor小波理论在目标特征提取和识别中的处理方法，有机地把二者结合起来。同时构建了用于灰度图像目标识别的Oabor小波神经网络模型，应用优化理论和自适应技术，使目标识别系统达到实时处理的目的。
本文研究的主要成果和创新点有：
(1)提出了一种适于多CPU目标识别系统改进的BP神经网络训练算法。主要是用变步长方法及输入向量归一化方法，对所选激励函数在理论上做了调整，最后，在多CPU的目标识别的神经网络系统上实现了该算法。
(2)基于多分辨分析原理，设计一种新的基于Gabor小波的多通道滤波器。该滤波器能对低质量的灰度图象目标进行特征提取，并有很好的鲁棒性。滤波器的中心频率是一个从低到高的范围，滤波器采用6方向，4尺度，对灰度图象直接进行小波变换，用Gabor小波变换系数的模的平均值和其标准方差来表示抽取的灰度图象目标的纹理特征，最后，把获得的小波特征输入到改进的BP神经网络分类器进行分类识别。
(3)提出了一种基于Log Gabor小波的低层次的图象分割、边缘特征
中国科学院博士学位论文:Gabor小波神经网络及其在灰度图象目标识别中的应用研究
提取的方法。主要是运用小波变换频域中的相位信息提取目标特征不变量，
这种不变量对目标的亮度、光线变化具有不变性。运用PC特征不变量特征
结合神经网络算法对灰度图象目标进行了识别，识别率明显的提高。
(4)提出了Gabor小波神经网络模型及其训练算法。主要从理论方面
对Gabor小波神经网络进行了分析，同时运用BP算法原理构建了一种适合
目标识别的Gabor小波神经网络训练算法。理论和仿真实验表明，Gabor小
波神经网络的收敛性和鲁棒性均明显优于BP网络。该算法应用于目标识别
时，不仅提高了识别的精度，而且克服了BP算法易陷入局部极小的缺陷。
(5)对多神经网络集成算法进行了初步的探讨研究，给出了一种用于
自动目标识别(ArR)的神经网络集成算法，并结合算法设计了一个基于多
CPU并行结构的多目标识别神经网络系统。该算法是运用不同规模和初始
条件下形成的同种类型的神经网络分类器。主要运用的是改进的BP神经网
络，并把每一个神经网络输出映射为后验概率，然后进行加权平均判决。最
后利用设计的神经网络系统进行了实时识别。
The dissertation is the study of Gabor wavelet neural network algorithm and its application in gray image target recognition. It is involved with the neural network theory, the wavelet analysis theory ,the wavelet neural network and their application technology of the target recognition. The key algorithms and technologies in the system realization are studied and discussed from the theory and practice , the aim of the research is the designing and realizing the neural networks target system based on many CPU .This thesis mainly consists of the following parts.
the actuality of the BP neural network algorithm, the wavelet transforms theory and Gabor wavelet neural network algorithm and their application of target recognition in the world are introduced. The problems are analyzed. The basic theory and the key technologies are expounded, the main toll route are gray image target recognition.
The multiplayer forward neural network and its training algorithm are thorough analyzed, Error back propagation algorithm is derived from the mathematic , the problem of BP algorithm is indicated .The improved BP algorithm with many target recognition is constructed.
The principle of Gabor filter is expounded. The expression of Gabor wavelet filter is presented. The multi channel Gabor filter is designed based on theory and practicality, the texture features of gray image target are extracted. The neural network recognizing algorithm based on multi channel Gabor filter feature is presented .
The define of phase congruence (PC) invariant is introduced. The mathematic process of deriving phase congruence (PC) invariant is discussed from theory ,the formula of PC invariant is presented.
The performance of log gabor wavelet is detailedly analyzed, the modified formula of PC invariant is presented by the log gabor wavelet. Finally.the new algorithm of neural network recognition based on the PC invariant is designed.
The three target representations are expatiated, that are the target representation based on the feature , based on the template and based Gabor wavelet neural network, the advantage and disadvantage are discussed. The construct method of Gabor wavelet neural network is expounded. The training algorithm of Gabor wavelet neural network is constructed,the theory analysis and the concrete step of algorithm are presented.
The mostly thought t are real time recognizing gray image target with Gabor wavelet neural networks algorithm. The train of thoughts are the forward neural networks (BP net) and Gabor wavelet are organically combined based on they were applied in target feature extraction and recognition. A model of Gabor wavelet neural network is constructed with automatic target recognition, the real time process aim is realized with the automatic target recognizing system applying optimizing theory and self-adapt technique. The neural networks ensemble algorithm is realized with neural network processing system designed based many CPU ,and the good impact is gained when it is applied target recognition .
Mostly research findings and innovations are focused on :
(1) The improved BP neural networks training algorithm was given out suit to many CPU target recognition system Principally with alter step and input vector , excitation function was adjusted from theory, finally this algorithm was realized in target recognizing system based on many CPU.
(2) According to multi resolve principle,a new multi channel filter based Gabor wavelet was designed. It can extract the feature of low quality gray image target, and had good robust. Its center frequency was the range from low frequency to high frequency, its orientation is 6 and scale is 4. Gary image was directly transformed by these wavelet filters, the feature of extracting gray image target was denoted by the coefficients of Gabor wavelet transform and its standard variance, the wavelet feature was input to the improved BP neural networks to classify.
(3) The method of image segment and edge feature extraction of low level was put f
对目前国际、国内BP神经网络算法、小波变换理论、Gabor小波神经网络算法以及在目标识别中的应用现状进行了综述，对存在的问题进行了剖析。同时以灰度图像目标识别为主线，阐述与论文直接相关的基本理论和关键技术。
详细分析了多层前馈型神经网络描述及训练算法机理，从数学的角度推导了误差逆传播算法(BP算法)，同时指出了BP算法存在的问题。构建了一种用于多目标识别的改进的BP算法。
阐述了Gabor滤波器原理。给出了Gabor小波滤波器的表达式。根据理论分析和实际需要，设计了多通道Gabor滤波器，提取了灰度图像目标纹理特征。相应地给出了一个基于多通道Gabor滤波器特征神经网络识别算法。
给出了相位叠加不变量pc的定义。从理论上详细探讨了用Gabor小波求解相位叠加不变量PC的数学过程，给出了相应的PC不变量的公式。
详细分析了log gabor函数性能，给出了用log gabor求相位叠加(PC)不变量的修正公式。最后，设计了基于相位叠加(PC)不变量的神经网络目标识别新算法。
阐述了三种不同的目标表示法，即基于特征的目标描述法、基于模板的目标描述法和基于Gabor小波神经网络目标描述法，同时分析了三种表示法的优缺点。详细阐述了Gabor小波神经网络构建方法，重点构建了Gabor小波神经网络的训练算法，给出了理论分析和算法的具体步骤。
本文的主要研究思想就是：运用并行的Gabor小波神经网络算法对灰度图像目标进行实时识别。研究的思路：根据前馈神经网络模型(BP网络)和Gabor小波理论在目标特征提取和识别中的处理方法，有机地把二者结合起来。同时构建了用于灰度图像目标识别的Oabor小波神经网络模型，应用优化理论和自适应技术，使目标识别系统达到实时处理的目的。
本文研究的主要成果和创新点有：
(1)提出了一种适于多CPU目标识别系统改进的BP神经网络训练算法。主要是用变步长方法及输入向量归一化方法，对所选激励函数在理论上做了调整，最后，在多CPU的目标识别的神经网络系统上实现了该算法。
(2)基于多分辨分析原理，设计一种新的基于Gabor小波的多通道滤波器。该滤波器能对低质量的灰度图象目标进行特征提取，并有很好的鲁棒性。滤波器的中心频率是一个从低到高的范围，滤波器采用6方向，4尺度，对灰度图象直接进行小波变换，用Gabor小波变换系数的模的平均值和其标准方差来表示抽取的灰度图象目标的纹理特征，最后，把获得的小波特征输入到改进的BP神经网络分类器进行分类识别。
(3)提出了一种基于Log Gabor小波的低层次的图象分割、边缘特征
中国科学院博士学位论文:Gabor小波神经网络及其在灰度图象目标识别中的应用研究
提取的方法。主要是运用小波变换频域中的相位信息提取目标特征不变量，
这种不变量对目标的亮度、光线变化具有不变性。运用PC特征不变量特征
结合神经网络算法对灰度图象目标进行了识别，识别率明显的提高。
(4)提出了Gabor小波神经网络模型及其训练算法。主要从理论方面
对Gabor小波神经网络进行了分析，同时运用BP算法原理构建了一种适合
目标识别的Gabor小波神经网络训练算法。理论和仿真实验表明，Gabor小
波神经网络的收敛性和鲁棒性均明显优于BP网络。该算法应用于目标识别
时，不仅提高了识别的精度，而且克服了BP算法易陷入局部极小的缺陷。
(5)对多神经网络集成算法进行了初步的探讨研究，给出了一种用于
自动目标识别(ArR)的神经网络集成算法，并结合算法设计了一个基于多
CPU并行结构的多目标识别神经网络系统。该算法是运用不同规模和初始
条件下形成的同种类型的神经网络分类器。主要运用的是改进的BP神经网
络，并把每一个神经网络输出映射为后验概率，然后进行加权平均判决。最
后利用设计的神经网络系统进行了实时识别。
The dissertation is the study of Gabor wavelet neural network algorithm and its application in gray image target recognition. It is involved with the neural network theory, the wavelet analysis theory ,the wavelet neural network and their application technology of the target recognition. The key algorithms and technologies in the system realization are studied and discussed from the theory and practice , the aim of the research is the designing and realizing the neural networks target system based on many CPU .This thesis mainly consists of the following parts.
the actuality of the BP neural network algorithm, the wavelet transforms theory and Gabor wavelet neural network algorithm and their application of target recognition in the world are introduced. The problems are analyzed. The basic theory and the key technologies are expounded, the main toll route are gray image target recognition.
The multiplayer forward neural network and its training algorithm are thorough analyzed, Error back propagation algorithm is derived from the mathematic , the problem of BP algorithm is indicated .The improved BP algorithm with many target recognition is constructed.
The principle of Gabor filter is expounded. The expression of Gabor wavelet filter is presented. The multi channel Gabor filter is designed based on theory and practicality, the texture features of gray image target are extracted. The neural network recognizing algorithm based on multi channel Gabor filter feature is presented .
The define of phase congruence (PC) invariant is introduced. The mathematic process of deriving phase congruence (PC) invariant is discussed from theory ,the formula of PC invariant is presented.
The performance of log gabor wavelet is detailedly analyzed, the modified formula of PC invariant is presented by the log gabor wavelet. Finally.the new algorithm of neural network recognition based on the PC invariant is designed.
The three target representations are expatiated, that are the target representation based on the feature , based on the template and based Gabor wavelet neural network, the advantage and disadvantage are discussed. The construct method of Gabor wavelet neural network is expounded. The training algorithm of Gabor wavelet neural network is constructed,the theory analysis and the concrete step of algorithm are presented.
The mostly thought t are real time recognizing gray image target with Gabor wavelet neural networks algorithm. The train of thoughts are the forward neural networks (BP net) and Gabor wavelet are organically combined based on they were applied in target feature extraction and recognition. A model of Gabor wavelet neural network is constructed with automatic target recognition, the real time process aim is realized with the automatic target recognizing system applying optimizing theory and self-adapt technique. The neural networks ensemble algorithm is realized with neural network processing system designed based many CPU ,and the good impact is gained when it is applied target recognition .
Mostly research findings and innovations are focused on :
(1) The improved BP neural networks training algorithm was given out suit to many CPU target recognition system Principally with alter step and input vector , excitation function was adjusted from theory, finally this algorithm was realized in target recognizing system based on many CPU.
(2) According to multi resolve principle,a new multi channel filter based Gabor wavelet was designed. It can extract the feature of low quality gray image target, and had good robust. Its center frequency was the range from low frequency to high frequency, its orientation is 6 and scale is 4. Gary image was directly transformed by these wavelet filters, the feature of extracting gray image target was denoted by the coefficients of Gabor wavelet transform and its standard variance, the wavelet feature was input to the improved BP neural networks to classify.
(3) The method of image segment and edge feature extraction of low level was put f
引文
[1]B.Bhanu et al ,Introduction to the special issue on automatic target detection and recognition[J].IEEE Trans. On Image Processing, 1997,6(1 ): 1-3.
[2] S Grossberg, H.Hawkins,A .Waxman. Introduction :1995 special issue automatic target recognition[J] .Neural Networks, 1995,(8)7:1003.
[3] Werbos P.J. Back propagation: past and future. Proc. IEEE Conf. Neural Networks, 1988, 1: 342-353.
[4] Rumelhart D.E.and McClelland J.L. Parallel Distributed Processing: Explorations in the Microstructure of Cognition[J], Vol. 1,Foundations MIT Press.Cambridge,MA, 1986。
[5] McClelland J. L. and Rumelhart D .E. Parallel Distributed Processing :Explorations in the Microstructure of Cognition[J],Vol.2,Foundations MIT Press.Cambridge,MA, 1986
[6] M.I.Miller, et al.Conditional-mean estimation via jump-diffusion processes in multiple target tracking/recognition[J]. IEEE transactions on Signal Processing, 1995, vol. 43(11): 2678-2690.
[7] Nait Charif Hammadi and Hideo Ito.a leraning algorithm for fault tolerant feed forward neural networks [J]. IEICE Trans. INF&SYST. ,1997, vol. E80-D (1) : 21-26.
[8] ruck thawonmas,et al. A cascade neural networks for blind signal extraction without spuriours equilibria[J].IEICE trans. Fundamentals,1998,vol.E81-A(9):1833-1845.
[9] P.V.S.Ponnapalli,et al. A formal selection and pruning algorithm for feedforward artificial neural network optimition[J]. IEEE transactions on Neural Networks, 1999, vol. 10 (4): 964-968.
[10] Kazuyuki Hara and Kenji Nakayama.Multi-frequency signal classfication by multiplayer neural networks and linear filter methods[J]. IEICE trans. Fundamentals, 1997,vol.E80-A(5):894-902.
[11] A.K Nandi and E.E.Azzouz.Modulation recognition using artificial neural networks[J] ,Signal Processing, 1997, vol. 56: 165-175.
[12] Dirk Husmerier and John G.Taylor.Neural Networks for predicting conditional probability densities:improved training scheme combining EM and RVFL[J].Neural Networks, 1998, vol. 11(1): 89-116.
[13] R.Hecht-Nielsen.Theory of the back-propagation neural network[J]. Proceeding of the International Joint Conference on Neural Networks, 1989,1:593-605.
[14] Cybenko.Approximation by superpositions of a sigmoidal function[J], Mathematics of Control,Signal and Systems,1999,3:303-314.
[15] Widrow B..30 years of adapative neural networks: perceptron, madaline, and backpropagation[J]. IEEE transactions on Neural Networks, 1990, vol. 1(9): 1415-1442
[16] 周礼臬，人工神经网络在生物医学中应用，中国神经网络首届学术大会，1990．12，北京。
[17] R.C.Eberhart, Neural Network PC Tools-A Practical Guide ,Academic Press Inc.,New York, 1990
[18] Y.Lee, Handwrtten Digit Recognition Using K Nearest-Neighbor, Radial-Basis Fuction,and Backpropagation Neural Networks[J] ,Neural Computation 3,pp.440-449,1991
[19] S.D Wang and C.C.Pan, A Neural Network Approach for Chinese Character Recognition[J], Proc.of IJCNN'90,Vol. Ⅱ ,pp.882-888.San Diego,1990.
[20] M.Mougeot,R.Azencott and B.Angeniol, Image Compression with Backpropagation:Improvement of the Visual Restoration Using Different Cost Function [J], Neural Networks Vol. 4, pp.467-476,1991.
[21] D.J.Burr, Experiments on Neural Net Recognition of Spoken and Written Text[J],IEEE
Trans.ASSP, Vol.36,No.7,pp. 1162-1168,1988.
[22] D.A.Pomerleau,Efficient on Neural Net Recognition of Spoken and Written Text[J],IEEE Trans.ASSP, Vol.36,No.7,pp. 1162-1168,1988.
[23] D.J.Sobajic and Y.-H Pao,Artificial Neural-Net Based Dynamic Security Assessment for Electric Power System[J],IEEE Trans.PWRS,Vol.4,No. 1,pp.220-228,Feb 1989.
[24] M.J.Damborg et al,Potential of Artificial Neural Networks in Power System Operation[J],Proc.of ISCAS'90,pp.2933-2937,1990.
[25] T.M.Peng et al,Conceptual Approach to the Application of Neural Network for Short-Term Load Forrecasting[J],Proc.of ISCAS'90,pp.2942-2945,1990.
[26] P.J.Werbos,Backpropagation Through Time :What It Does and How to Do It[J], Proc.of IEEE, Vol. 78, No. 10,pp.1550-1560,1990.
[27] K.A.McCrae,D.W.Ruck,S.K.Rogers,and M.E.Oxley, Color image segmentation[J], Applications of Artificial Neural Networks V, in Proc.SPIE 2243,306-315(1994)
[28] H.G.Stark and G.P.Laufkoetter, Image indexing and content based access to databases of medial images with wavelets[J], in Wavelet Applications in signal and Image Processing Ⅲ,Proc.SPIE 2569,790-799(1995)
[29] J.S.Smoklin, Wavelet feature extraction for image pattern recognition[J], in Hybrid Image and Signal Processing V, Proc.SPIE 2751,110-121(1996)
[30] W.Wang,Z.Bao,Q.Meng, G.M.Flachs,and J.B.Jordan, Hand recognition by wavelet transform and neural network[J], Signal Processing ,Sensor Fusion, and Target Recognition Ⅳ, I.Kadar and V.Libby, Eds.,Proc.SPIE 2484 347-353(1995)
[31] L.Chan ans N.Nasrabadi, Modular wavelet-based vector quantization for automatic target reconition[J]. in Proc.IEEE/SICE/RSJ International conference on Multisensor Fusion and Integration for Intelligent systems,pp.462-469(1996)
[32] Jia, Jiancheng; Pattern classification of RGB color images using a BP neural network classifier [J], AA(Nanyang Technological Univ.) Publication: Proc. SPIE Vol. 1989, p. 248-256
[33] Pascal; Savaria, Yvon; Daniel, Herve; Cantin, Marc-Andre; Blaquiere, Yves;Hardware/software codesign of a fuzzy ART neural clusterer: the benefits of configurable computing Poire[J], AA(Ecole Polytechnique de Montreal) AD(Univ. du Quebec/Montreal) Publication: Proc. SPIE Vol. 3526, p. 90-96
[34]. Kong, Ling; Liu, Chunping; Shen, Peihua; Xia, Deshen; Recognition of lymph node metastasis malignancy tumor cells based on the multilayer error back-propagation (BP) neural network[J], AA(Nanjing Univ. of Science and Technology) Publication: Proc. SPIE Vol. 4555, p. 57-64,
[35]. k Jiang, Ming; Zhang, Guilin; Chen, Qijie; Matching method applying neural network[J], Publication: Proc. SPIE Vol. 4552, p. 63-68,
[36] Fei, Gou; Yu, Yinglin; P Modified sigma-pi BP network with self-feedback and its application in time seriesroc[J]. SPIE Vol. 2243, p. 508-515,
[37] S. N.; Paulraj, M.; Nithyanandam, Mathiyazhagan;Certain improvements in back propagation procedure for pattern identification Sivanandam[J], Publication: Proc. SPIE Vol. 4055, p. 79-89
[38] 杜正春等。前馈神经网络的一种有效学习算法[J]．电子学报，1995，vol．23(8)：57-61。
[39] Raffaele Parisi.et al.A generalized learning paradigm exploiting the structure of feedforward neural networks[J].IEEE transactions on neural netwoks, 1996, vol. 7 (6): 1450-1459.
[40] Xiao-Hu and Guo-an Chen .Efficient back-propagetion learning using optimal learning rate and
momentum[J].Neural Networks, 1997,vol. 10(3):517-527.
[41]穆文全等。基于遗传算法和BP算法的多层感知机杂交训练算法[J]。电子科学学刊，1997，vol．19(2)：190-194。
[42]John. J. McKeown , et al. Some numerical aspects of the training networks.Proceedings [J] IEEE, 1996,vol.84(10):1562-1568.
[43]S.C.NG, S.H.LEUNG and A.LUK.Fast convergent generalized backpropagation algorithm with constant learning rate[J].Neural Processing Letters, 1999,9:13-23.
[44]Jiri Sima.Back-propagation is not efficient[J]. Neural Networks, 1996, vol.9(6): 1017-1023.
[45]Charles W.Lee.Training feedforward neural networks:an algorithm giving improved generalization[J].Neural Networks, 1997,vol. 10(1):62-68.
[46]叶世伟，史忠植。前向神经网络动态学习[J]。电子学报，1998。VOL。26(11)：140-144。
[47]冯伟．高大启。前向神经网络的隐节点修剪算法[J]。电子学报，1997，VOL。25(5)：114-115。
[48]叶东毅。前向多层神经网络隐层节点设计的一个学习算法[J]。电子学报，1997，VOL，35(11)：126-128。
[49]Osamu Fujita.Statistical estimation of the number of hidden units for feedforward neural networks[J]. Neural Networks, 1998,vol. 11(11):851-859.
[50]Roy S&Shynk J J..Analysis of the momentum LMS algorithm factor[J]. IEEE ISCAS, USA, 1992, 5.
[51]杨大力，刘泽民。多层前向神经网络中BP算法的误调分析及其改进算法[J]。电子学报，1995，VOL。23(1)：117-120。
[52]Georg Thimm et al.The interchangeablility of learning rate and gain in backpropagation neural network[J]. Neural Computation, 1996,8:451-460.。
[53]石俊等。基于自适应噪声的鲁棒BP算法及前馈网的鲁棒性分析[J]。系统工程与电子技术，1998，VOL。(4)：47-51。
[54]Ingrid Daubechies.The wavelet transform,time-frequency localization and signal analysis[J].IEEE transactions on Information Theory. 1990,vol.36(5):961-1003.
[55]Nurgun Erdol and Filiz Basbug.Wavelet transform based adaptive filters:analysis and new results[J]. IEEE transactions on Signal Processing, 1996,vol.44(9):2163-2171.
[56]Milos I. Dorosiovacki and H.Fan.Wavelet based linear sysytem modeling and adaptive filtering[J] .IEEE transactions on Signal Processing, 1996,vol.44(5): 1156-1167.
[57]Ming-jun Lai.On the digital filter associated with Daubechies'wavelet[J].IEEE transactions on Signal Processing, 1995,vol.43(9):2203-2205.
[58]丁宏等。采用小波变换岁短数据信号的谱估计方法[J]。电子学报，1997，VOL。25(1)：11-13。
[59]Philip L. et al .A B-wavelet-based noise-reduction algorithm[J] .IEEE transactions on Signal Processing, 1996,vol.44(55):1279-1284.
[60]Wim Sweldens .Wavelets:what next? [J]。Proceedings of the IEEE,1996,VOL.84(4):680-685.
[61]李正东等，利用小波变换进行目标识别的方法研究[J]。光学 精密工程，2001 VOL。9(3)：216-219。
[62]S,Mallat,A theory for multiresolution signal decomposition: the wavelet representation[J].IEEE Trans.Pattern Anal.Mach.Intell. 11.674-693(1989).
[63]P.L.Ainsleigh and C.K.Chui.A B-wavelet based noise reduction algorithm[J]. IEEE trans. Signal Process. 1996,44(5), 1279-1284
[64]D.L.Donoho,De-Noising by Soft-Thresholding[J], IEEE Trans.Inf.Theory 1995,41(3):613-627.
[65]Shen Dinggang ,Horace H Sip.Discriiminative wavelet shape descriptors for recognition of 2-D patterns [J].pattern Recognition, 1999,32:151-165
[66]Tai Sing Lee. Image representation Using 2D Gabor wavelet[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 1996, 18(10): 959-971
[67]Volker Kruger. Wavelet subspace method for real-time face tracking[C]. Proc. Pattern Recognition, 23rd DAGM Symposium, Munich, Germany, 2001.
[68]Volker Kruger, A. Happe and G. Sommer. Affine Real-time tracking using Gabor wavelet networks[C]. ICCV'99 Workshop Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems (RATFG-RTS'99), Corfu, Greece, Sept. 26-27, 1999.
[69]Daubechies I. The wavelets teansform, time-frequency localization and signal analysis[J]. IEEE Trans. On Info.Theory. 1990,36 (5):961-1005
[70]Daugman G. Complete discrete 2-Dtransforms by neural networks for image analysis and compression[J].IEEE Trans.on ASSP, 1988,ASSP-36(7):1169-1179
[71]Denk T. etal.Combining neural networks and the wavelet transform for image compression[J]. Proc. IEEE ICASSP. Minnesota, USA: 1993, 1:637-640.
[72]Szu H.,Yang X,Telfer B,et al . Neural Network and wavelet transform as preprocessor for the scale-invariant data classification[J].Phys.Rev.E, 1993,48(2):1497-1501
[73]Kalayci T, Ozdamar O,Erdol N.The Use of wavelet transform as preprocessor for the neural network detection of EEG spikes[J].Proc.IEEE Southeast Conference, 1994,1-3
[74]Q.H.Zhang and Albert Benveniste.Wavelet network[J].IEEE transactions on Neural Network, 1992,vol.3(6):889-898
[75]Y.C.Pati and P.S. Krishnaprasad.Anlysis and synthesis of feedforward neural networks using discrete affine wavelet transformation[J]. IEEE transactions on Neural Network, 1993, vol. 4(1): 73-85.
[76]张帮礼等。小波神经网络的构造机器算法的鲁棒性分析[J]。重庆大学学报，1995，VOL。18．(6)：88-95
[77]Jun Zhang ,eta al.Wavelet neural networks for function learning[J] .IEEE transactions on Signal Processing, 1995,vol.43(6):1485-1497
[78]B.Delyon,A.Juditaky, A.Benveniste.Accuracy analysis for wavelet approximations[J].IEEE transactions on Neural Network, 1995,vol.6(2):332-348.
[79]He. Shichun and He Zhenya.Application of recurrent wavelet neural networks to the digital communications channel blind equalization. Journal of China Institute of Communications [J], 1997,vol. 18(3):65-69]
[80]Huang Dexian, et al.Application research of wavelet neural networks in process control[J]．清华大学学报，1999，vol．39(1)：91-94．]
[81]徐晓霞等。基于正交最小二乘算法的小波神经网络[J]。电子学报，1998，VOL。26(10)：115-117。
[82]何振亚。用于信号逼近的自适应时延小波神经网络[J]。电子科学学刊，1998，VOL。20(5)：604-610。
[83]张艳宁等。用于船舶噪声分类的局域自适应子波神经网络分类器[J]。系统工程与电子技术，1998，6：21-25。
[84]刘芳等，一种基于子波的模糊推理神经网络及其学习算法研究[J]。电子学报，1999，VOL。27(2)：32-36
[85]Lee S.H. Optical implementations of digital algorithms for pattern recognition [J]. Opt. Engng. 1986, 25(1):69-75.
[86]Freeman.M.O.Salch.B.E.A.Optical location of cancroids of non-overlapping objects [J]. App. Opt. 1987: 26(14/15): 2752-2759
[87]Hun H.Pan J.K. Optical/digital invariant recognition of two-dimensional patterns with straight lines [J].
Opt. Engng.1996,35(4):997-1002.
[88]申金嫒，刘钥，陈戍，郭鹏毅，张延炘。用于实时识别三维物体的级联神经网络模型的优化方法[J]。光学学报第 21卷，第1期，2001年1月：49-53。
[89]阎平凡，张长水编著．人工神经网络和模拟进化计算[M]．清华大学出版社．2001年2月第2次印刷．
[90]Ken Tabb. Nell Davey, et al. The recognition and analysis of animate objects using neural networks and active contour models[J]. Neurocomputing 43 (2002): 145-172
[91]周建华，白钊．光度信号的频域模式识别方法[J]．光学精密工程．2002，Vol10，No．4：346-353
[92]刘文耀等编著。光电图象处理[M]。电子工业出版社。2002年11月第1版。
[93]A.C.Bovic,M.Clark, and W.S.Geisler, Multichannel texture analysis using localized spatial filters[J]. IEEE Trans.Pattem Analysis and Machine Intelligence,vol. 12.no 1 .pp55-73 Jan 1990.
[94]B.S.Manjunath,C.Shekhar and R.Chetlappa, A new approach to image feature detection with applications[J]. Pattern Recognition Apr. 1996
[95]潘诗林，周源华，用干目标识别的二维傅里叶小波描述子[J]．红外与激光工程，第31卷，第 4期,2002年8月：290=293
[96]Xing.Wu,Bir Bhanu,Gabor wavelet representation for 3-D object recognition[J]. IEEE Trans. Image Processing1997 1;6(1):47-64
[97]王学文，丁小青，刘长松，基于gabor变换的高鲁棒汉字识别新方法[J]。电子学报 2002年第9期：1317-1322
[98]容观澳，计算机图象处理[M]，清华大学出版社，2000年第1版
[99]KENNETH。R。CASTLEMAN著，朱志刚等译，数字图象处理[M]，电子工业出版社，1998年第1版
[100]S.Marcelja, Mathematical Description of the Responses of Simple Cortical Cells[J], J.Opt. Soc. A., vol. 70, 1980,1297-1230.
[101]K.K.De Valois,Spatial Vision,Annual Review of Physiology, 1984.
[102]D.H.Hubel and T.N.Wiesel,Receptive Fields,Binocular Interaction and Functional Aracitecure in the Cat's Visual Cortex[J],J.Physiology, vol. 160,1962,106-154.
[103]M.R.Turner, Texture Discrimination by Gabor Functions[J],Biol.Cybern,vol.55,1986,71-82
[104]A.K.Jain and F.Farrokhnia, Unsupervised Texture Segmentation Using Gabor Filters[J],Pattern Recognition,vol.24,no. 12,1991, 1167-1186.
[105]H.Knustsson and G.H.Granlund,Texture analysis using two-dimensional quadrature filters[J],IEEE Workshop Comput.Arcthitect.Pattern Anal.Image Database Management, Oct. 1983,12-14.
[106]A.C.Bovik,Analysis of Multiehannel Narrow-Band Filters for Image Texture Segmentation[J],IEEE Trans.on SP, vol.39,1991,2925-2043.
[107]F.Farrokhnia and A.K.Jain A Multi—Channel Filtering approach to Texture Segmentation[J], Proc. Of IEEE CVPR'91,June 3-6,1991,364-370
[108]J.J.Kuliknowski,S.Marcelja, and P.Obisshop,Theory of spatial position and spatial frequency relations in the receptive fields of simple cells in the visual cortex[J], Biol. Cybern., vol. 43, 1982,187-198.
[109]J.G.Daugman,Two-dimensional spectral analysis of cortical receptive field profiles[J],Vision Res.,vol.20,1980,847-856.
[110]张军，戚飞虎。基于子波分解的多通道神经网络纹理分割方法[J]，红外与毫米波学报，1998，17(1)，54-60。
[111]M.J.Bastiaans,Ssampling theorem for the complex spectrogram and Gabor's expansion of a signal in Gaussian elementary signals [J],Opt.Eng.,vol.20,1981,594-598.
[112]M.J.Bastiaans,Gabor's signal expansion and degrees of freesom of a signal[J], Opt. Acta, vol. 29, 1982, 1223-1229.
[113]M.K.Hu.Visual pattern recognition by moment invariants[J].IRE Transations in Information Theory, IT-8:179-187,1962.
[114]孙即祥等著。模式识别中的特征提取与计算机视觉不变量[M]北京：．国防工业出版社，2001年9月第一版。
[115]M.C.Morrone and R.A.Owens.Feature detection from local energy[J].Pattern Recognition Letters,6:303-313,1987
[116]S.Venkatesh and R.A.Owens. An energy feature detection scheme[C]. Singapore: In The International Conference on Image Processing,pages 553-557,1989
[117]M.C.Morrone and R.A.Owens.Feature detection from local energy[J]. Pattern Recognition Letters,6:303-313,1987.
[118]J.G.Daugman.Uncertainty relation for resolution in space,spatial frequency, and orientation optimized by two-dimensional visual cortical filters[J].J of The Optical Society of America A, Jul., 1985,2(7): 1160-1169
[119]M.A.Webster, R.L.De Valois.Relationshio between spatial -frequency and orientation tunning of strate-cortex cells[J].J of The Optical Society of America A, Jul.,1985,2(7): 1124-1132
[120]S.J.Anderson,D.C.Burr, Spatial and temporal selectivity of the human motion detection system.Vision Research,25(8): 1147-1154,1985
[121]STEPHANE MALLAT著，杨力华等译。信号处理的小波导引[M]。机械工业出版社2002年9月第一版
[122]Chen Tianping,Chen Hong Approximation capability to function of several variables ,nonlinear functionals , and operators by radial basis function neural network.. [J] IEEE Trans.on NN, 1995,6(4):904～910.
[123]Delyon B,Juditsky A and Benveniste A. Accuracy analysis for wavelet approximations.[J] IEEE Trans. On NN, 1995,6(2):332～348.
[124]Kreinovich V, Sirisaengtaksin O and Cabren S. Wavelet neural networks are asymptotically optimal approximators for functions of one variable.[C] Florida, USA: Proceeding of IEEEICNN' 94,1994.1:299～304.
[125]Zhang Q,Benveniste A. Wavelet networks. [J] IEEE Trans. on NN, 1992, 3(11): 889～898.
[126]Pati YC,Krishnaprasad PS.Analysis and synthesis of feedforward neural network using discrete affine wavelet., Baskshi [J].IEEE Trans. On NN,1993,4(1):73～75.
[127]Baskshi BR, Stephanopoulous G. Wave-net: a multiresolution, hierarchical neural network with localized learning. [J] American Institute Chemical Engineer Journal, 1993,39(1):57～81.
[128]Zhang Jun,Walter G, MiaoY, etal. Wavelet neural networks for function learning. [J] IEEE Trans.on SP,1995,43(6):1485～1497
[129]Shi Y, Zhang X.-D. Gabor atom networks for signal classification with application in radar target recognition[J]. IEEE Trans. Signal Processing ,2001,49:2994-3004.
[130]杨行峻，郑君里．人工神经网络[M]．北京：高等教育出版社，1992．
[131]R Hecbt-Nielsen. Kolmogorov' s mapping neural network existence theorem[C].Proceedings of International Conference on Neural Networks. New York:IEEE Press, 1987.3:11-14.
[132]K Hornik, M Stinchcombe, H White. Multilayer feedforward neural networks are universal approximators[J]. Neural Networks, 1989,2(2): 359-366.
[133]Stinchcombe M, White H. Universal Approximation Using Feedforward Networks with Non-Sigmoidal Hidden Layers Activation Function[J]. Proc. 1989 International Joint Conference on Neural Networks, San Diego; SOS Printing, 1989(1): 613-617
[134]Y X Zhang, Z Song, Y Lin. Invariant image encoding in a neural network for target recognition[A].CD-ROM of Proc IJCNN' 99[C].Cd(D.G. Brown),Washington D.C. USA, 1999.
[135]盛积德，张延炘，常胜江，陈戍，前馈神经网络中隐层神经元的研究[J]。光电子 激光 第 12卷第6期2001年6月620-622
[136]Hansen L. Salamon T. Neural networks ensemble[J].IEEE Trans on Pattern Analysis and Machine Intelligence 1990 12(10) 993-1001.
[137]Hansen L.K. Lissberg L. Salamon P. Ensemble methods for handwritten digit recognition[J].In:Proc of the IEEE Workshop on Neural Networks for Signal Processing .Copenhagen Denmark. 1992,333-342.
[138]Kirkland J. Squad-based expert models for closing diphthong recognition[C]. In:Proc of the 2ndNew Zealand Int Two-stream Conf.on Artificial Neural Networks and Expert Systerms Dunedin,New Zealand, 1995,302-305
[139]周志华 皇甫杰等．基于神经网络集成的多视角人脸识别[J]．计算机研究与发展2001 10 1204-1220
[140]Wolpert D. Stacked generalizationS[J]. Neural Networks , 1993 5(2):241-259.
[141]Homik K M, Stinchcombe M, White H. Multilayer Fecdforward Networks Are Universal Approximators[J]. Neural Networks, 1989, 2(2): 359～366.
[142]Kearns M, Valiant L G. Learning Boolean Formulae or Factoring. Technical Report TR-1488, Cambridge, MA: Havard University Aiken Computation Laboratory, 1988.
[143]Schapire R E. The Strength of Weak Learnability[J]. Machine Learning, 1990, 5(2): 197～227.
[144]Sollich P, Krogh A. Learning with Ensembles: How Over-Fitting Can Be Useful. In: Touretzky D, Mozer M, Hasselmo M eds. Advances in Neural Information Processing Systems (Volume 8), Cambridge, MA: MIT Press, 1996, 190～196.
[145]Opitz D, Maclin R. Popular Ensemble Methods: An Empirical Study[J]. Journal of Artificial Intelligence Research, 1999, 11: 169～198.
[146]Cooper L N. Hybrid Neural Network Architectures: Equilibrium Systems That Pay Attention. In: Mammone R J, Zeevi Y Y eds. Neural Networks: Theory and Applications, San Diego, CA: Academic Press, 1991, 81～96.
[147]Perrone M P, Coopler L N. When Networks Disagree: Ensemble Method for Neural Networks. In: Mammone R J ed. Artificial Neural Networks for Speech and Vision, London: Chapman-Hall, 1993, 126～142.
[148]Opitz D, Shavlik J. Actively Searching for An Effective Neural Network Ensemble[J]. Connection Science, 1996, 8(3-4): 337～353.
[149]Zhang X, Mesirov J, Waltz D. Hybrid System for Protein Secondary Structure Prediction[J]. Journal of Molecular Biology, 1992, 225: 1049～1063.
[150]Rost B, Sander C. Prediction of Protein Secondary Structure at Better Than 70% Accuracy[J]. Journal of Molecular Biology, 1993, 232: 584～599.
[151]Breiman L. Bagging Predictors. [J] Machine Learning, 1996, 24(2): 123～140.
[152]Freund Y. Boosting a Weak Algorithm by Majority[J]. Information and Computation, 1995, 121(2): 256～285.
[153]Freund Y, Schapire R E. A Decision-Theoretie Generalization of On-Line Learning and An Application to Boosting[J]. Journal of Computer and System Sciences, 1997, 55(1): 119～139.
[154]Opitz D W, Shavlik J W. Generating Accurate and Diverse Members of A Neural Network Ensemble. In: Touretzky D, Mozer M, Hasselmo Meds. Advances in Neural Information Processing Systems (Volume 8), Cambridge, MA: MIT Press, 1996, 535～541.
[155]Yao X, Liu Y. Making Use of Population Information in Evolutionary Artificial Neural Networks[J]. IEEE Transactions on Systems. Man and Cybernetics- Part B: Cybernetics, 1998, 28(3): 417～425.
[156]Hampshire J, Waibel A. A Novel Objective Function for Improved Phoneme Recognition Using Time-Delay Neural Networks[J]. IEEE Transactions on Neural Networks, 1990, 1(2): 216～228.
[157]Cherkauer K J. Human Expert Level Performance on A Scientific Image Analysis Task by A System Using Combined Artificial Neural Networks. In: Proceedings of the 13th AAAI Workshop on Integrating Multiple Learned Models for Improving and Scaling Machine Learning Algorithms, Menlo Part, CA: AAAI Press, 1996, 15～21.
[158]Maclin R, Shavlik J W. Combining the Predictions of Multiple Classifiers: Using Competitive Learning to Initialize Neural Networks. In: Proceedings of the 14th International Joint Conference on Artificial Intelligence, San Mateo, CA: Morgan Kaufmann, 1995, 524～530.
[159]鲁湛 丁晓青．基于分类器判决可靠度估计的最优线性集成方法．计算机学报 2002 2：890-895
[160]Lars Kai Hansen, Peter Salamon, "Neural Network Ensembles", IEEE Tran. on PAMI, vol. 12, no.10, pp993-1001, 1990
[161]Roberto Battiti, Anna Maria Colla, "Democracy in Neural Nets: Voting Schemes for Classification", Neural Networks, vol 7, no 4, pp691-707, 1994
[162]Michael D. Richard, et al, "Neural Network Classifiers Estimate Bayesian a posteriori Probabilities", Neural Computation, pp461-483, 1991.3
[163]Tin Kam Ho, J. J. Hull, and Sargur N. Srihari, "Decision Combination in Multiple Classifier Systems", IEEE Trans. on PAMI, Vol. 16, No.1, Jan 1994, pp.66-75
[164]Louisa Lam, et. al, "Optimal Combinations of Pattern Classifiers", Pattern Recog. Letters, vol. 16, pp945-954, 1995
[2] S Grossberg, H.Hawkins,A .Waxman. Introduction :1995 special issue automatic target recognition[J] .Neural Networks, 1995,(8)7:1003.
[3] Werbos P.J. Back propagation: past and future. Proc. IEEE Conf. Neural Networks, 1988, 1: 342-353.
[4] Rumelhart D.E.and McClelland J.L. Parallel Distributed Processing: Explorations in the Microstructure of Cognition[J], Vol. 1,Foundations MIT Press.Cambridge,MA, 1986。
[5] McClelland J. L. and Rumelhart D .E. Parallel Distributed Processing :Explorations in the Microstructure of Cognition[J],Vol.2,Foundations MIT Press.Cambridge,MA, 1986
[6] M.I.Miller, et al.Conditional-mean estimation via jump-diffusion processes in multiple target tracking/recognition[J]. IEEE transactions on Signal Processing, 1995, vol. 43(11): 2678-2690.
[7] Nait Charif Hammadi and Hideo Ito.a leraning algorithm for fault tolerant feed forward neural networks [J]. IEICE Trans. INF&SYST. ,1997, vol. E80-D (1) : 21-26.
[8] ruck thawonmas,et al. A cascade neural networks for blind signal extraction without spuriours equilibria[J].IEICE trans. Fundamentals,1998,vol.E81-A(9):1833-1845.
[9] P.V.S.Ponnapalli,et al. A formal selection and pruning algorithm for feedforward artificial neural network optimition[J]. IEEE transactions on Neural Networks, 1999, vol. 10 (4): 964-968.
[10] Kazuyuki Hara and Kenji Nakayama.Multi-frequency signal classfication by multiplayer neural networks and linear filter methods[J]. IEICE trans. Fundamentals, 1997,vol.E80-A(5):894-902.
[11] A.K Nandi and E.E.Azzouz.Modulation recognition using artificial neural networks[J] ,Signal Processing, 1997, vol. 56: 165-175.
[12] Dirk Husmerier and John G.Taylor.Neural Networks for predicting conditional probability densities:improved training scheme combining EM and RVFL[J].Neural Networks, 1998, vol. 11(1): 89-116.
[13] R.Hecht-Nielsen.Theory of the back-propagation neural network[J]. Proceeding of the International Joint Conference on Neural Networks, 1989,1:593-605.
[14] Cybenko.Approximation by superpositions of a sigmoidal function[J], Mathematics of Control,Signal and Systems,1999,3:303-314.
[15] Widrow B..30 years of adapative neural networks: perceptron, madaline, and backpropagation[J]. IEEE transactions on Neural Networks, 1990, vol. 1(9): 1415-1442
[16] 周礼臬，人工神经网络在生物医学中应用，中国神经网络首届学术大会，1990．12，北京。
[17] R.C.Eberhart, Neural Network PC Tools-A Practical Guide ,Academic Press Inc.,New York, 1990
[18] Y.Lee, Handwrtten Digit Recognition Using K Nearest-Neighbor, Radial-Basis Fuction,and Backpropagation Neural Networks[J] ,Neural Computation 3,pp.440-449,1991
[19] S.D Wang and C.C.Pan, A Neural Network Approach for Chinese Character Recognition[J], Proc.of IJCNN'90,Vol. Ⅱ ,pp.882-888.San Diego,1990.
[20] M.Mougeot,R.Azencott and B.Angeniol, Image Compression with Backpropagation:Improvement of the Visual Restoration Using Different Cost Function [J], Neural Networks Vol. 4, pp.467-476,1991.
[21] D.J.Burr, Experiments on Neural Net Recognition of Spoken and Written Text[J],IEEE
Trans.ASSP, Vol.36,No.7,pp. 1162-1168,1988.
[22] D.A.Pomerleau,Efficient on Neural Net Recognition of Spoken and Written Text[J],IEEE Trans.ASSP, Vol.36,No.7,pp. 1162-1168,1988.
[23] D.J.Sobajic and Y.-H Pao,Artificial Neural-Net Based Dynamic Security Assessment for Electric Power System[J],IEEE Trans.PWRS,Vol.4,No. 1,pp.220-228,Feb 1989.
[24] M.J.Damborg et al,Potential of Artificial Neural Networks in Power System Operation[J],Proc.of ISCAS'90,pp.2933-2937,1990.
[25] T.M.Peng et al,Conceptual Approach to the Application of Neural Network for Short-Term Load Forrecasting[J],Proc.of ISCAS'90,pp.2942-2945,1990.
[26] P.J.Werbos,Backpropagation Through Time :What It Does and How to Do It[J], Proc.of IEEE, Vol. 78, No. 10,pp.1550-1560,1990.
[27] K.A.McCrae,D.W.Ruck,S.K.Rogers,and M.E.Oxley, Color image segmentation[J], Applications of Artificial Neural Networks V, in Proc.SPIE 2243,306-315(1994)
[28] H.G.Stark and G.P.Laufkoetter, Image indexing and content based access to databases of medial images with wavelets[J], in Wavelet Applications in signal and Image Processing Ⅲ,Proc.SPIE 2569,790-799(1995)
[29] J.S.Smoklin, Wavelet feature extraction for image pattern recognition[J], in Hybrid Image and Signal Processing V, Proc.SPIE 2751,110-121(1996)
[30] W.Wang,Z.Bao,Q.Meng, G.M.Flachs,and J.B.Jordan, Hand recognition by wavelet transform and neural network[J], Signal Processing ,Sensor Fusion, and Target Recognition Ⅳ, I.Kadar and V.Libby, Eds.,Proc.SPIE 2484 347-353(1995)
[31] L.Chan ans N.Nasrabadi, Modular wavelet-based vector quantization for automatic target reconition[J]. in Proc.IEEE/SICE/RSJ International conference on Multisensor Fusion and Integration for Intelligent systems,pp.462-469(1996)
[32] Jia, Jiancheng; Pattern classification of RGB color images using a BP neural network classifier [J], AA(Nanyang Technological Univ.) Publication: Proc. SPIE Vol. 1989, p. 248-256
[33] Pascal; Savaria, Yvon; Daniel, Herve; Cantin, Marc-Andre; Blaquiere, Yves;Hardware/software codesign of a fuzzy ART neural clusterer: the benefits of configurable computing Poire[J], AA(Ecole Polytechnique de Montreal) AD(Univ. du Quebec/Montreal) Publication: Proc. SPIE Vol. 3526, p. 90-96
[34]. Kong, Ling; Liu, Chunping; Shen, Peihua; Xia, Deshen; Recognition of lymph node metastasis malignancy tumor cells based on the multilayer error back-propagation (BP) neural network[J], AA(Nanjing Univ. of Science and Technology) Publication: Proc. SPIE Vol. 4555, p. 57-64,
[35]. k Jiang, Ming; Zhang, Guilin; Chen, Qijie; Matching method applying neural network[J], Publication: Proc. SPIE Vol. 4552, p. 63-68,
[36] Fei, Gou; Yu, Yinglin; P Modified sigma-pi BP network with self-feedback and its application in time seriesroc[J]. SPIE Vol. 2243, p. 508-515,
[37] S. N.; Paulraj, M.; Nithyanandam, Mathiyazhagan;Certain improvements in back propagation procedure for pattern identification Sivanandam[J], Publication: Proc. SPIE Vol. 4055, p. 79-89
[38] 杜正春等。前馈神经网络的一种有效学习算法[J]．电子学报，1995，vol．23(8)：57-61。
[39] Raffaele Parisi.et al.A generalized learning paradigm exploiting the structure of feedforward neural networks[J].IEEE transactions on neural netwoks, 1996, vol. 7 (6): 1450-1459.
[40] Xiao-Hu and Guo-an Chen .Efficient back-propagetion learning using optimal learning rate and
momentum[J].Neural Networks, 1997,vol. 10(3):517-527.
[41]穆文全等。基于遗传算法和BP算法的多层感知机杂交训练算法[J]。电子科学学刊，1997，vol．19(2)：190-194。
[42]John. J. McKeown , et al. Some numerical aspects of the training networks.Proceedings [J] IEEE, 1996,vol.84(10):1562-1568.
[43]S.C.NG, S.H.LEUNG and A.LUK.Fast convergent generalized backpropagation algorithm with constant learning rate[J].Neural Processing Letters, 1999,9:13-23.
[44]Jiri Sima.Back-propagation is not efficient[J]. Neural Networks, 1996, vol.9(6): 1017-1023.
[45]Charles W.Lee.Training feedforward neural networks:an algorithm giving improved generalization[J].Neural Networks, 1997,vol. 10(1):62-68.
[46]叶世伟，史忠植。前向神经网络动态学习[J]。电子学报，1998。VOL。26(11)：140-144。
[47]冯伟．高大启。前向神经网络的隐节点修剪算法[J]。电子学报，1997，VOL。25(5)：114-115。
[48]叶东毅。前向多层神经网络隐层节点设计的一个学习算法[J]。电子学报，1997，VOL，35(11)：126-128。
[49]Osamu Fujita.Statistical estimation of the number of hidden units for feedforward neural networks[J]. Neural Networks, 1998,vol. 11(11):851-859.
[50]Roy S&Shynk J J..Analysis of the momentum LMS algorithm factor[J]. IEEE ISCAS, USA, 1992, 5.
[51]杨大力，刘泽民。多层前向神经网络中BP算法的误调分析及其改进算法[J]。电子学报，1995，VOL。23(1)：117-120。
[52]Georg Thimm et al.The interchangeablility of learning rate and gain in backpropagation neural network[J]. Neural Computation, 1996,8:451-460.。
[53]石俊等。基于自适应噪声的鲁棒BP算法及前馈网的鲁棒性分析[J]。系统工程与电子技术，1998，VOL。(4)：47-51。
[54]Ingrid Daubechies.The wavelet transform,time-frequency localization and signal analysis[J].IEEE transactions on Information Theory. 1990,vol.36(5):961-1003.
[55]Nurgun Erdol and Filiz Basbug.Wavelet transform based adaptive filters:analysis and new results[J]. IEEE transactions on Signal Processing, 1996,vol.44(9):2163-2171.
[56]Milos I. Dorosiovacki and H.Fan.Wavelet based linear sysytem modeling and adaptive filtering[J] .IEEE transactions on Signal Processing, 1996,vol.44(5): 1156-1167.
[57]Ming-jun Lai.On the digital filter associated with Daubechies'wavelet[J].IEEE transactions on Signal Processing, 1995,vol.43(9):2203-2205.
[58]丁宏等。采用小波变换岁短数据信号的谱估计方法[J]。电子学报，1997，VOL。25(1)：11-13。
[59]Philip L. et al .A B-wavelet-based noise-reduction algorithm[J] .IEEE transactions on Signal Processing, 1996,vol.44(55):1279-1284.
[60]Wim Sweldens .Wavelets:what next? [J]。Proceedings of the IEEE,1996,VOL.84(4):680-685.
[61]李正东等，利用小波变换进行目标识别的方法研究[J]。光学 精密工程，2001 VOL。9(3)：216-219。
[62]S,Mallat,A theory for multiresolution signal decomposition: the wavelet representation[J].IEEE Trans.Pattern Anal.Mach.Intell. 11.674-693(1989).
[63]P.L.Ainsleigh and C.K.Chui.A B-wavelet based noise reduction algorithm[J]. IEEE trans. Signal Process. 1996,44(5), 1279-1284
[64]D.L.Donoho,De-Noising by Soft-Thresholding[J], IEEE Trans.Inf.Theory 1995,41(3):613-627.
[65]Shen Dinggang ,Horace H Sip.Discriiminative wavelet shape descriptors for recognition of 2-D patterns [J].pattern Recognition, 1999,32:151-165
[66]Tai Sing Lee. Image representation Using 2D Gabor wavelet[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 1996, 18(10): 959-971
[67]Volker Kruger. Wavelet subspace method for real-time face tracking[C]. Proc. Pattern Recognition, 23rd DAGM Symposium, Munich, Germany, 2001.
[68]Volker Kruger, A. Happe and G. Sommer. Affine Real-time tracking using Gabor wavelet networks[C]. ICCV'99 Workshop Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems (RATFG-RTS'99), Corfu, Greece, Sept. 26-27, 1999.
[69]Daubechies I. The wavelets teansform, time-frequency localization and signal analysis[J]. IEEE Trans. On Info.Theory. 1990,36 (5):961-1005
[70]Daugman G. Complete discrete 2-Dtransforms by neural networks for image analysis and compression[J].IEEE Trans.on ASSP, 1988,ASSP-36(7):1169-1179
[71]Denk T. etal.Combining neural networks and the wavelet transform for image compression[J]. Proc. IEEE ICASSP. Minnesota, USA: 1993, 1:637-640.
[72]Szu H.,Yang X,Telfer B,et al . Neural Network and wavelet transform as preprocessor for the scale-invariant data classification[J].Phys.Rev.E, 1993,48(2):1497-1501
[73]Kalayci T, Ozdamar O,Erdol N.The Use of wavelet transform as preprocessor for the neural network detection of EEG spikes[J].Proc.IEEE Southeast Conference, 1994,1-3
[74]Q.H.Zhang and Albert Benveniste.Wavelet network[J].IEEE transactions on Neural Network, 1992,vol.3(6):889-898
[75]Y.C.Pati and P.S. Krishnaprasad.Anlysis and synthesis of feedforward neural networks using discrete affine wavelet transformation[J]. IEEE transactions on Neural Network, 1993, vol. 4(1): 73-85.
[76]张帮礼等。小波神经网络的构造机器算法的鲁棒性分析[J]。重庆大学学报，1995，VOL。18．(6)：88-95
[77]Jun Zhang ,eta al.Wavelet neural networks for function learning[J] .IEEE transactions on Signal Processing, 1995,vol.43(6):1485-1497
[78]B.Delyon,A.Juditaky, A.Benveniste.Accuracy analysis for wavelet approximations[J].IEEE transactions on Neural Network, 1995,vol.6(2):332-348.
[79]He. Shichun and He Zhenya.Application of recurrent wavelet neural networks to the digital communications channel blind equalization. Journal of China Institute of Communications [J], 1997,vol. 18(3):65-69]
[80]Huang Dexian, et al.Application research of wavelet neural networks in process control[J]．清华大学学报，1999，vol．39(1)：91-94．]
[81]徐晓霞等。基于正交最小二乘算法的小波神经网络[J]。电子学报，1998，VOL。26(10)：115-117。
[82]何振亚。用于信号逼近的自适应时延小波神经网络[J]。电子科学学刊，1998，VOL。20(5)：604-610。
[83]张艳宁等。用于船舶噪声分类的局域自适应子波神经网络分类器[J]。系统工程与电子技术，1998，6：21-25。
[84]刘芳等，一种基于子波的模糊推理神经网络及其学习算法研究[J]。电子学报，1999，VOL。27(2)：32-36
[85]Lee S.H. Optical implementations of digital algorithms for pattern recognition [J]. Opt. Engng. 1986, 25(1):69-75.
[86]Freeman.M.O.Salch.B.E.A.Optical location of cancroids of non-overlapping objects [J]. App. Opt. 1987: 26(14/15): 2752-2759
[87]Hun H.Pan J.K. Optical/digital invariant recognition of two-dimensional patterns with straight lines [J].
Opt. Engng.1996,35(4):997-1002.
[88]申金嫒，刘钥，陈戍，郭鹏毅，张延炘。用于实时识别三维物体的级联神经网络模型的优化方法[J]。光学学报第 21卷，第1期，2001年1月：49-53。
[89]阎平凡，张长水编著．人工神经网络和模拟进化计算[M]．清华大学出版社．2001年2月第2次印刷．
[90]Ken Tabb. Nell Davey, et al. The recognition and analysis of animate objects using neural networks and active contour models[J]. Neurocomputing 43 (2002): 145-172
[91]周建华，白钊．光度信号的频域模式识别方法[J]．光学精密工程．2002，Vol10，No．4：346-353
[92]刘文耀等编著。光电图象处理[M]。电子工业出版社。2002年11月第1版。
[93]A.C.Bovic,M.Clark, and W.S.Geisler, Multichannel texture analysis using localized spatial filters[J]. IEEE Trans.Pattem Analysis and Machine Intelligence,vol. 12.no 1 .pp55-73 Jan 1990.
[94]B.S.Manjunath,C.Shekhar and R.Chetlappa, A new approach to image feature detection with applications[J]. Pattern Recognition Apr. 1996
[95]潘诗林，周源华，用干目标识别的二维傅里叶小波描述子[J]．红外与激光工程，第31卷，第 4期,2002年8月：290=293
[96]Xing.Wu,Bir Bhanu,Gabor wavelet representation for 3-D object recognition[J]. IEEE Trans. Image Processing1997 1;6(1):47-64
[97]王学文，丁小青，刘长松，基于gabor变换的高鲁棒汉字识别新方法[J]。电子学报 2002年第9期：1317-1322
[98]容观澳，计算机图象处理[M]，清华大学出版社，2000年第1版
[99]KENNETH。R。CASTLEMAN著，朱志刚等译，数字图象处理[M]，电子工业出版社，1998年第1版
[100]S.Marcelja, Mathematical Description of the Responses of Simple Cortical Cells[J], J.Opt. Soc. A., vol. 70, 1980,1297-1230.
[101]K.K.De Valois,Spatial Vision,Annual Review of Physiology, 1984.
[102]D.H.Hubel and T.N.Wiesel,Receptive Fields,Binocular Interaction and Functional Aracitecure in the Cat's Visual Cortex[J],J.Physiology, vol. 160,1962,106-154.
[103]M.R.Turner, Texture Discrimination by Gabor Functions[J],Biol.Cybern,vol.55,1986,71-82
[104]A.K.Jain and F.Farrokhnia, Unsupervised Texture Segmentation Using Gabor Filters[J],Pattern Recognition,vol.24,no. 12,1991, 1167-1186.
[105]H.Knustsson and G.H.Granlund,Texture analysis using two-dimensional quadrature filters[J],IEEE Workshop Comput.Arcthitect.Pattern Anal.Image Database Management, Oct. 1983,12-14.
[106]A.C.Bovik,Analysis of Multiehannel Narrow-Band Filters for Image Texture Segmentation[J],IEEE Trans.on SP, vol.39,1991,2925-2043.
[107]F.Farrokhnia and A.K.Jain A Multi—Channel Filtering approach to Texture Segmentation[J], Proc. Of IEEE CVPR'91,June 3-6,1991,364-370
[108]J.J.Kuliknowski,S.Marcelja, and P.Obisshop,Theory of spatial position and spatial frequency relations in the receptive fields of simple cells in the visual cortex[J], Biol. Cybern., vol. 43, 1982,187-198.
[109]J.G.Daugman,Two-dimensional spectral analysis of cortical receptive field profiles[J],Vision Res.,vol.20,1980,847-856.
[110]张军，戚飞虎。基于子波分解的多通道神经网络纹理分割方法[J]，红外与毫米波学报，1998，17(1)，54-60。
[111]M.J.Bastiaans,Ssampling theorem for the complex spectrogram and Gabor's expansion of a signal in Gaussian elementary signals [J],Opt.Eng.,vol.20,1981,594-598.
[112]M.J.Bastiaans,Gabor's signal expansion and degrees of freesom of a signal[J], Opt. Acta, vol. 29, 1982, 1223-1229.
[113]M.K.Hu.Visual pattern recognition by moment invariants[J].IRE Transations in Information Theory, IT-8:179-187,1962.
[114]孙即祥等著。模式识别中的特征提取与计算机视觉不变量[M]北京：．国防工业出版社，2001年9月第一版。
[115]M.C.Morrone and R.A.Owens.Feature detection from local energy[J].Pattern Recognition Letters,6:303-313,1987
[116]S.Venkatesh and R.A.Owens. An energy feature detection scheme[C]. Singapore: In The International Conference on Image Processing,pages 553-557,1989
[117]M.C.Morrone and R.A.Owens.Feature detection from local energy[J]. Pattern Recognition Letters,6:303-313,1987.
[118]J.G.Daugman.Uncertainty relation for resolution in space,spatial frequency, and orientation optimized by two-dimensional visual cortical filters[J].J of The Optical Society of America A, Jul., 1985,2(7): 1160-1169
[119]M.A.Webster, R.L.De Valois.Relationshio between spatial -frequency and orientation tunning of strate-cortex cells[J].J of The Optical Society of America A, Jul.,1985,2(7): 1124-1132
[120]S.J.Anderson,D.C.Burr, Spatial and temporal selectivity of the human motion detection system.Vision Research,25(8): 1147-1154,1985
[121]STEPHANE MALLAT著，杨力华等译。信号处理的小波导引[M]。机械工业出版社2002年9月第一版
[122]Chen Tianping,Chen Hong Approximation capability to function of several variables ,nonlinear functionals , and operators by radial basis function neural network.. [J] IEEE Trans.on NN, 1995,6(4):904～910.
[123]Delyon B,Juditsky A and Benveniste A. Accuracy analysis for wavelet approximations.[J] IEEE Trans. On NN, 1995,6(2):332～348.
[124]Kreinovich V, Sirisaengtaksin O and Cabren S. Wavelet neural networks are asymptotically optimal approximators for functions of one variable.[C] Florida, USA: Proceeding of IEEEICNN' 94,1994.1:299～304.
[125]Zhang Q,Benveniste A. Wavelet networks. [J] IEEE Trans. on NN, 1992, 3(11): 889～898.
[126]Pati YC,Krishnaprasad PS.Analysis and synthesis of feedforward neural network using discrete affine wavelet., Baskshi [J].IEEE Trans. On NN,1993,4(1):73～75.
[127]Baskshi BR, Stephanopoulous G. Wave-net: a multiresolution, hierarchical neural network with localized learning. [J] American Institute Chemical Engineer Journal, 1993,39(1):57～81.
[128]Zhang Jun,Walter G, MiaoY, etal. Wavelet neural networks for function learning. [J] IEEE Trans.on SP,1995,43(6):1485～1497
[129]Shi Y, Zhang X.-D. Gabor atom networks for signal classification with application in radar target recognition[J]. IEEE Trans. Signal Processing ,2001,49:2994-3004.
[130]杨行峻，郑君里．人工神经网络[M]．北京：高等教育出版社，1992．
[131]R Hecbt-Nielsen. Kolmogorov' s mapping neural network existence theorem[C].Proceedings of International Conference on Neural Networks. New York:IEEE Press, 1987.3:11-14.
[132]K Hornik, M Stinchcombe, H White. Multilayer feedforward neural networks are universal approximators[J]. Neural Networks, 1989,2(2): 359-366.
[133]Stinchcombe M, White H. Universal Approximation Using Feedforward Networks with Non-Sigmoidal Hidden Layers Activation Function[J]. Proc. 1989 International Joint Conference on Neural Networks, San Diego; SOS Printing, 1989(1): 613-617
[134]Y X Zhang, Z Song, Y Lin. Invariant image encoding in a neural network for target recognition[A].CD-ROM of Proc IJCNN' 99[C].Cd(D.G. Brown),Washington D.C. USA, 1999.
[135]盛积德，张延炘，常胜江，陈戍，前馈神经网络中隐层神经元的研究[J]。光电子 激光 第 12卷第6期2001年6月620-622
[136]Hansen L. Salamon T. Neural networks ensemble[J].IEEE Trans on Pattern Analysis and Machine Intelligence 1990 12(10) 993-1001.
[137]Hansen L.K. Lissberg L. Salamon P. Ensemble methods for handwritten digit recognition[J].In:Proc of the IEEE Workshop on Neural Networks for Signal Processing .Copenhagen Denmark. 1992,333-342.
[138]Kirkland J. Squad-based expert models for closing diphthong recognition[C]. In:Proc of the 2ndNew Zealand Int Two-stream Conf.on Artificial Neural Networks and Expert Systerms Dunedin,New Zealand, 1995,302-305
[139]周志华 皇甫杰等．基于神经网络集成的多视角人脸识别[J]．计算机研究与发展2001 10 1204-1220
[140]Wolpert D. Stacked generalizationS[J]. Neural Networks , 1993 5(2):241-259.
[141]Homik K M, Stinchcombe M, White H. Multilayer Fecdforward Networks Are Universal Approximators[J]. Neural Networks, 1989, 2(2): 359～366.
[142]Kearns M, Valiant L G. Learning Boolean Formulae or Factoring. Technical Report TR-1488, Cambridge, MA: Havard University Aiken Computation Laboratory, 1988.
[143]Schapire R E. The Strength of Weak Learnability[J]. Machine Learning, 1990, 5(2): 197～227.
[144]Sollich P, Krogh A. Learning with Ensembles: How Over-Fitting Can Be Useful. In: Touretzky D, Mozer M, Hasselmo M eds. Advances in Neural Information Processing Systems (Volume 8), Cambridge, MA: MIT Press, 1996, 190～196.
[145]Opitz D, Maclin R. Popular Ensemble Methods: An Empirical Study[J]. Journal of Artificial Intelligence Research, 1999, 11: 169～198.
[146]Cooper L N. Hybrid Neural Network Architectures: Equilibrium Systems That Pay Attention. In: Mammone R J, Zeevi Y Y eds. Neural Networks: Theory and Applications, San Diego, CA: Academic Press, 1991, 81～96.
[147]Perrone M P, Coopler L N. When Networks Disagree: Ensemble Method for Neural Networks. In: Mammone R J ed. Artificial Neural Networks for Speech and Vision, London: Chapman-Hall, 1993, 126～142.
[148]Opitz D, Shavlik J. Actively Searching for An Effective Neural Network Ensemble[J]. Connection Science, 1996, 8(3-4): 337～353.
[149]Zhang X, Mesirov J, Waltz D. Hybrid System for Protein Secondary Structure Prediction[J]. Journal of Molecular Biology, 1992, 225: 1049～1063.
[150]Rost B, Sander C. Prediction of Protein Secondary Structure at Better Than 70% Accuracy[J]. Journal of Molecular Biology, 1993, 232: 584～599.
[151]Breiman L. Bagging Predictors. [J] Machine Learning, 1996, 24(2): 123～140.
[152]Freund Y. Boosting a Weak Algorithm by Majority[J]. Information and Computation, 1995, 121(2): 256～285.
[153]Freund Y, Schapire R E. A Decision-Theoretie Generalization of On-Line Learning and An Application to Boosting[J]. Journal of Computer and System Sciences, 1997, 55(1): 119～139.
[154]Opitz D W, Shavlik J W. Generating Accurate and Diverse Members of A Neural Network Ensemble. In: Touretzky D, Mozer M, Hasselmo Meds. Advances in Neural Information Processing Systems (Volume 8), Cambridge, MA: MIT Press, 1996, 535～541.
[155]Yao X, Liu Y. Making Use of Population Information in Evolutionary Artificial Neural Networks[J]. IEEE Transactions on Systems. Man and Cybernetics- Part B: Cybernetics, 1998, 28(3): 417～425.
[156]Hampshire J, Waibel A. A Novel Objective Function for Improved Phoneme Recognition Using Time-Delay Neural Networks[J]. IEEE Transactions on Neural Networks, 1990, 1(2): 216～228.
[157]Cherkauer K J. Human Expert Level Performance on A Scientific Image Analysis Task by A System Using Combined Artificial Neural Networks. In: Proceedings of the 13th AAAI Workshop on Integrating Multiple Learned Models for Improving and Scaling Machine Learning Algorithms, Menlo Part, CA: AAAI Press, 1996, 15～21.
[158]Maclin R, Shavlik J W. Combining the Predictions of Multiple Classifiers: Using Competitive Learning to Initialize Neural Networks. In: Proceedings of the 14th International Joint Conference on Artificial Intelligence, San Mateo, CA: Morgan Kaufmann, 1995, 524～530.
[159]鲁湛 丁晓青．基于分类器判决可靠度估计的最优线性集成方法．计算机学报 2002 2：890-895
[160]Lars Kai Hansen, Peter Salamon, "Neural Network Ensembles", IEEE Tran. on PAMI, vol. 12, no.10, pp993-1001, 1990
[161]Roberto Battiti, Anna Maria Colla, "Democracy in Neural Nets: Voting Schemes for Classification", Neural Networks, vol 7, no 4, pp691-707, 1994
[162]Michael D. Richard, et al, "Neural Network Classifiers Estimate Bayesian a posteriori Probabilities", Neural Computation, pp461-483, 1991.3
[163]Tin Kam Ho, J. J. Hull, and Sargur N. Srihari, "Decision Combination in Multiple Classifier Systems", IEEE Trans. on PAMI, Vol. 16, No.1, Jan 1994, pp.66-75
[164]Louisa Lam, et. al, "Optimal Combinations of Pattern Classifiers", Pattern Recog. Letters, vol. 16, pp945-954, 1995