成像制导中的图像预处理及目标识别技术研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
作为一种自主式智能精确制导技术,红外成像制导在现代空战中发挥着越来越重要的作用,而自动目标识别则是其中的核心技术。但目前目标识别技术还有许多困难需要解决,而且这些困难分别存在于识别系统中的预处理、特征提取和分类识别等环节中。在此背景下,本文以红外成像制导的图像处理分析和目标识别为主线,针对各个环节所存在的困难,系统研究了目标识别系统中的图像滤波、目标分割、二维目标特征提取、三维目标特征提取和分类识别等问题。论文所取得的主要研究成果如下:
     1.在图像滤波技术中,现有滤波方法存在无法同时实现噪声去除与边缘保持的问题。针对这个问题,本文提出了一种各向异性高斯分段滤波器,在对滤波器的参数进行估计的基础上,使滤波器仅在有效邻域滤波。但是各向异性高斯分段滤波器又存在着计算量较大的缺点,为此本文又提出了一种分段矩滤波来解决这些问题,它在快速局部矩卷积算法和边缘阶梯模型的基础上实现了快速的有效邻域滤波,计算机仿真结果表明该方法能在有效去除噪声的同时较好地保持边缘。
     2.图像分割是目标特征提取的基础,针对现有自适应阈值分割方法在非均匀背景下不能有效分割出目标的问题,提出了一种基于背景拟合的图像自适应分割方法,首先根据噪声统计信息估计出图像的噪声均方差,然后据此对图像背景进行了光顺限制的曲面拟合。通过与其它自适应阈值方法分割结果的比较表明,在非均匀背景下本文所提出的这种背景拟合分割方法具有较好的目标分割效果。
     3.在实际目标识别系统中,为了提高计算速度并简化系统常采用二维目标识别方法。本文针对常用的二维RST(旋转、尺度、位移)不变特征维数小且对噪声往往不稳定、而Trace变换较稳定但计算量过大的问题,提出了一种极坐标投影矩,它可以通过增减不变函数来调整不变量的特征维数。仿真实验结果表明这种不变量在尺度、旋转及噪声变化情况下仍具有很强稳定性,而分类实验则表明其具有良好的分类效果。
     4.针对二维目标识别系统中常用的仿射不变特征存在着特征维数较少且不变性不稳定,而规格化轮廓虽稳定但匹配过程复杂等缺点,提出了一种仿射投影矩特征,这种特征通过规格化和尺度归一化方法实现了对仿射变换的不变性,而且其特征维数的大小同样可以通过改变不变函数的数量来调整。计算机仿真结果验证了这种仿射投影矩对仿射变换的稳定性,并且分类实验表明了这种仿射投影矩的良好分类效果。
     5.针对三维目标识别系统中常用的正交矩没有分析特征之间相关性的情况,本文通过将K-L(Karhunen-Loeve)变换引入正交矩得到了具有最小相关性的最优正交识别矩——K-L傅立叶矩,但是K-L变换存在着正交基不固定、且对多类别数据识别效果不佳的缺点。为此,本文基于余弦变换是一阶马尔可夫过程中对K-L的最优近似的理论,提出了用余弦傅立叶矩来提取特征的方法,通过将余弦傅立叶矩分解为独立的余弦变换和傅立叶变换的方法,从而提高了计算速度。此外理论分析和实验仿真表明,余弦傅立叶矩的不变性对RST变换具有很强的稳定性,并且这种余弦傅立叶矩对于分类识别也有较好的效果。
     6.在分类算法中,支持向量机因能对样本进行结构风险最小化的分类而得到了广泛的应用。但现有多分类支持向量机方法在处理目标类别较多的情况时,存在着分类器训练速度慢、分类器较多、决策结构复杂且不稳定的问题。针对这种情况,本文提出了一种基于风险分析的最小随机风险多分类方法,当使用同类型的支持向量机对目标进行分类时,决策导向非循环图结构下的系统随机结构风险最小,计算机仿真结果表明其具有良好的分类效果。此外,将余弦傅立叶矩和支持向量机结合,用于对三维目标识别也取得了较高的识别精度。
As an autonomous intelligent accurate guidance, infrared imaging guidance is playing an increasingly important role in air war. In this system, targets recognition is key to guidance processing. But there are many problems need to be solved in targets recognition and these problems lie in different parts, such as preprocessing, feature extraction and classification. Under this background, this paper is organized on infrared image analyzing and targets recognition. To solve these problems, the image filtering, target segment, feature extraction of 2D and 3D targets recognition are investigated systemically in this dissertation. The important contributions and creative achievements are summarized as follows:
     1. As the existent filters haven't good effect in denoising and preserving edge, an anisotropic piecewise gaussian filter has been put forward. In this method, the parameters of filter are estimated to make it smooth only in effective neighbor. But the computation amount of this anisotropic piecewise gaussian filter is large, so a piecewise moment filter is developed in this dissertation. Based on fast convolution and stair edge model, the moment filter can make a fast smoothing in effective neighbor. The experiment result shows that this new method has better capability than other common methods in suppressing noise and preserving edge.
     2. Image segment is the base of feature extraction, but the standard segmentation methods are ineffective for extracting target in nonuniform background. To solve the problem, a surface fitting segmentation method is proposed. Firstly, a statistical model of noise has been used to estimate the variance of noise. Then the fairing constrained surface fitting is made for the image background. The experimental result shows that the new method has better capability than other common methods in removing background and extracting target.
     3. To increase the speed of computation and simplify the system, 2D targets recognition system is often used. In this dissertation, a polar-projection moment is proposed to solve the problems that the invariant features aren't stable and the computational amount of trace transform is large. In the polar-projection moment, dimension of these features can be adapted by the number of invariant functions. The simulation shows that these features have mighty stability when the target have scale, rotational and noise changes and classification experiment demonstrated that these features have good effect in classification.
     4. Furthermore, affine invadant features also have the problems that invariant features aren't stable and the matching process of normalized contour is complex. To solve these problems, a affine-projection moment is proposed. These features can realize invariance by normalization and scale average. After that, dimension of features also can be adjusted by the number of invariant functions. Simulation explains that these features are stable and the classification experiment displays that the data can be classified well by this affine projection moment.
     5. To get the least information redundancy of orthogonal moment, a K-L fourier moment is put forward by involving K-Ltransform to get the best orthogonal moment in recognition. But the K-L orthogonal bases are not fixed orthogonal bases and it is not good at recognizing the multi-classification data, so a cosine fourier moment is proposed. Because the cosine transform is the best approximation to K-L transform in one order markov process, the cosine fourier moment is the suboptimal orthogonal moment. As the cosine fourier moment can be divided into independent cosine transform and fourier transform, the speed of computation can be increased. The analysis in theory and experiments show that this feature can obtain stable invariance and classification ability.
     6. To reduce the excessive sub-classifier and complex structure of SVM multi-classification, a minimal random risk method is proposed by analyzing the risk of structure. When the same SVM classifiers are used in classification, the random structural risk of system is minimized. The experiment presents that it can get better result than DAG method and it's speed is high. Finally in the experiment, the recognition system composed of cosine fourier moment and SVM gets higher classification accuracy in processing 3D target.
引文
[1] 王美华.不变量理论在模式识别中的应用.燕山大学硕士学位论文,2004
    [2] 徐伟捷.CCD成像制导自动目标识别系统的设计及实现.西北工业大学硕士学位论文,2000
    [3] 张建.灰建模理论及其在目标识别中的应用.西北工业大学硕士学位论文,2002
    [4] 陈春玉.目标识别技术的现状与发展.声学技术,1999,18(4):185-188
    [5] 汪勇旭.图像目标识别方法研究.西北工业大学硕士学位论文,2003
    [6] 李春明.军用飞机雷达与红外隐身技术综述.现代防御技术,1995,(2):38-48
    [7] 陈泳颜.便携式地空导弹技术及发展.中国航天,1998(7):35-38
    [8] 李补莲.自动目标识别(ATR)技术发展述评.现代防御技术,2000,28(2):10-20
    [9] 马晓旭.中远距空空导弹武器系统技术.电光与控制,1996,(1):1-13
    [10] 李立坤.新一代空空导弹关键技术和发展战略.航空科学技术,2003,(3):25-28
    [11] 李全良.新世纪的空空导弹.中国航天,2001,(12):30-34
    [12] 鞠传文,杨秀珍,王超勇.空空导弹攻击防区外目标研究.电光与控制,2002,9(3):12-14
    [13] 高阳,李言俊,张科.人眼视觉仿生在光电目标信息检测、估计和跟踪技术中的应用.航空兵器,2006,(1):22-25
    [14] 张山,周凤岐.智能凝视模型在红外成像制导(ATR)中的应用.弹箭与制导学报,1998,(1):25-28
    [15] 吴晗平,易新建.现阶段末制导红外成像跟踪系统性能提高的技术途径.应用光学,2003,24(4):20-22
    [16] 方斌,陈天如.空空导弹红外成像制导关键技术分析.红外技术,2003,25(4):45-48
    [17] 方有培,汪立萍.红外成像制导武器现状及其对抗.航天电子对抗,2004,(2):60-64
    [18] 施德恒.红外成像导引头及其成像制导武器.光机电信息,2002,(8):34-38
    [19] 施德恒,黄宜军.红外成像导引头发展综述.激光与红外,1997,27(6):351-354
    [20] 刘永昌,吴鹏,张黎.迈向新世纪的智能化红外成像寻的制导技术.红外技术,2002,24(2):1-5
    [21] 卢福刚.红外图像目标识别与跟踪方法研究.西北工业大学博士学位论文,1998
    [22] 涂建平,彭应宁.用于空空导弹弹道终端的ATR系统和算法.光电工程,2003,30(1):24-28
    [23] 郁文贤,郭桂蓉.ATR的研究现状和发展趋势.系统工程与电子技术,1994,(6):25-32
    [24] 李忠辉,李军,李辉等.红外精确制导技术的现状及发展趋势.长春光学精密机械学院学报,2000,23(4):19-24
    [25] 杨树谦.精确制导技术发展现状与展望.航天控制,2004,22(4):17-20
    [26] 舒海燕.图像目标识别技术的研究与应用.西北工业大学硕士学位论文,2002
    [27] 高峰.二维数字图像处理中的非线性滤波技术.光子学报,1995,24(1):48~56
    [28] 吴建华,李迟生,周卫星.中值滤波与均值滤波的去噪性能比较.南昌大学学报(工科版),1998,20(1):32~35
    [29] Z. C. Lin and Q. Y. Shi. An anisotropic diffusion PDE for noise reduction and thin edge preservation. International Conference on Image Analysis and Processing, 1999: 102-107
    [30] C. A. Z. Barcelos, M. Boaventura, E. C. Silva, et al. A well-balanced flow equation for noise removal and edge detection. IEEE Transactions on Image Processing, 2003, 12(7): 751-763
    [31] Deng G, Cahill L W. An adaptive gaussian filter for noise reduction and edge detection. Nuclear Science Symposium and Medical Imaging Conference. IEEE Conference Record, 1994, (3): 1615-1619
    [32] Tan, H. L., Gelfand, S. B. and Delp, E. J. A comparative cost function approach to edge detection. IEEE Transaction on Systems, Man, and Cybernetics, 1989, 19: 1337-1349
    [33] Tan, H. L., Gelfand, S. B. and Delp, E. J. A cost minimization approach to edge detection using simulated annealing. IEEE Transaction on Pattern Analysis and Machine Intelligence, 1991, 14: 3-18
    [34] Bhandarkar, S. M., Zhang, Y., and Potter, W. D. An edge detection technique using genetic algorithm-based optimization. Pattern Recognition, 1994, 27: 1159-1180
    [35] 朱菊华,杨新,李俊,施鹏飞.基于纹理分析的保细节平滑滤波器.中国图象图形学报,2001,6(A版)(11):1058-1064
    [36] 田村秀行编著,金喜子,乔双译.计算机图像处理.北京:科学出版社.2004
    [37] Qiu p. Discontinuous regression surface fitting. The Annals of Statistics, 1998, 26: 2218-2245
    [38] Qiu P. A nonparametrie procedure to detect jumps in regression surface. Journal of Computational and Graphical Statistics, 2002, 11: 799-822
    [39] 王红梅.多传感器图像机器预处理技术研究.西北工业大学博士学位论文,2005
    [40] 董立菊.图像阈值化技术的综述、分类及评价.沈阳大学学报,2004,16(4):8-11
    [41] 王爱民,沈兰荪.图像分割研究综述.测控技术,2000,19(5):2-6
    [42] 王茜倩,彭中,刘莉.一种基于自适应阈值的图像分割算法.北京理工大学学报,2003,23(4):521-524
    [43] 杨波,叶俊勇,汪同庆,彭健.低对比度图像的自适应阈值化算法.传感器技术.2002,21(6):34-36
    [44] 付小宁,殷世民,吴志鹏,刘上乾.红外图像的动态阈值分割.光电工程,2002,29(6):57-60
    [45] 桑农,张天序,曹治国.基于边缘约束的红外目标图像松弛分割技术.电子学报,2002,30(7):1027-1031
    [46] Iwao Sekita, Takio Kurita, Nabih Abdelmalek. A Thresholding Method Using the Mixture of Normal Density Functions. Image Processing and Neural Networks International Symposium, 13-16 April 1994, (1): 304-307
    [47] 蔡勇,陈念年,李波.超声医学图像的自动分割.兰州交通大学学报(自然科学版).2004,23(4):61-64
    [48] 刘秀兰,马丹,刘兵,俞痒.一种新的快速动态闽值图像分割算法.北京工业大学学报,1999,25(2):92-96
    [49] 景云华,董才林,杨扬,李笑歌.复杂背景下的阈值插值方法.计算机工程,2003,29(17):5-6
    [50] llya Blayvas, Alfred Bruckstein and Ron Kimmel. Efficient Computation of Adaptive Threshold Surfaces for Image Binarization. Computer Vision and Pattern Recognition Proceedings of the 2001 IEEE Computer Society Conference, 2001, 1: 737-742
    [51] J R Parker, C Jennings, A G Salkauskas. Thresholding Using an Illumination Model Document Analysis and Recognition, Proceedings of the Second International Conference, 20-22 Oct, 1993: 270-273
    [52] 涂其远,吴建华,万国金.动态阈值结合全局闽值对图像进行分割[J].南昌大学学报(工科版),2002,24(1):37-40
    [53] Mark Hewish. Focal-Plane Arrays Revolutionize Thermal Imaging. International DefenceReview, 1993, 26(4): 307-311.
    [54] 吴军辉,都竹香,张玉竹.红外成像系统图像空间噪声分析与估计.红外技术,2001,23(3):19-22
    [55] 贾云德.机器视觉.北京:科学出版社,2000
    [56] 王耀明. 图像的矩函数:原理、算法及应用.上海:华东理工大学出版社,2002
    [57] 陈燕新,戚飞虎.一种新的适合于目标识别的仿射不变量.红外与毫米波学报,1999,18(1):61-66
    [58] 李言俊,张科.视觉仿生成像制导技术及应用.北京:国防工业出版社,2006
    [59] Ling J, Bovik A C. Smoothing low-SNR molecular images via anisotropic median-diffusion. IEEE Trans on Medical Imaging, 2002, 21(4): 377-384
    [60] 徐军.红外图像中弱小目标检测技术研究.西安:西安电子科技大学博士学位论文,2003
    [6l] 魏宏建,任浩,沈同圣.红外图像探测器噪声的数值模拟方法.航天电子对抗,2002,(6):33-35
    [62] 王学伟,沈同圣,周晓东.红外热像仿真中传感器的模拟方法研究.半导体光电,2004,25(4):317-319
    [63] 范宏波,李志广,杨帆 线阵红外焦平面成像系统中失效元检测及补偿的计算机模拟研究红外技术,2002,24(6):54-57
    [64] 王怀野,张科,李言俊.各向异性滤波在红外图像处理中的应用.红外与毫米波学报,2005,24(2):109-113
    [65] Jean-Bernard Martens. The Hermite Transform-Theory. IEEE Transactions on Acoustics Speech and Signal Processing. 1990, 38(9): 1595-1606
    [66] Shedf Makram-Ebeid and Benoit Mory. Scale-Space Image Analysis Based on Hermite Polynomials Theory. L. D. Griffin and M. Lillholm(Eds): Scale-Space 2003, LNCS2695: 57-71
    [67] 纪震,牟轩沁,蔡元龙.基于自适应方向滤波的左心室轮廓提取.中国图象图形学报,1999,4(A版)(8):649-654
    [68] 项静恬,史久恩.非线性系统中数据处理的统计方法.北京:科学出版社,2000
    [69] 穆振伟.多元正态分布均值向量有变点时当前均值的估计.烟台大学学报,1996,(3):4-10
    [70] 刘亚威,杨丹,张小洪.基于空间矩的亚像素边缘定位技术的研究.计算机应用,2003,23(2):47-49
    [71] 邹宁,柳健,周曼丽.基于Zemike矩的边缘检测方法.华中理工大学学报,1999,27(5):1-2
    [72] 刘力双,张铫,卢慧卿等.图像的快速亚像素边缘检测方法.光电子·激光,2005,16(8):993-996
    [73] 刘变莲.亚像素边缘检测技术的研究.电脑开发与应用,2005,18(11):54-55
    [74] 杨振亚,王淑仙,王成道.白适应图像边缘检测算法.计算机应用,2003,23(5):15-17
    [75] 张永宏,胡德金,张凯,徐俊杰.基于灰度矩的CCD图像亚像素边缘检测算法研究.光学技术,2004,30(6):693-695
    [76] 丁兴号,邓善熙,杨永跃等.基于空间矩和Zemike矩的亚像素边缘检测.应用科学学报2004,22(2):191-194
    [77] 冯卫东.工业零件形状尺寸图像检测技术研究.长沙:国防科学技术大学硕士学位论文,2002.
    [78] 巴本冬,吴晓娟,王磊.一种快速医学图像亚像素边缘检测方法.山东大学学报(工学版),2005,35(5):63-67
    [79] 章毓晋.图像理解与计算机视觉.北京:清华大学出版社,2000
    [80] Gongzhu Hu and Xing P. Cao. Subpixel edge detector using expectation of first-order derivatives. SPIE: Image Processing Algorithms and Techniques Ⅲ, 1992, 1657: 415-425
    [81] Stephen K. Park, Mark R. ldema. Algorithms for subpixel edge reconstruction. SPIE: Visual Information Processing V, 1996, 2753: 98-109
    [82] Ohtani, K. and Baba, M. A fast edge location measurement with subpixel accura-cy using a CCD image. Instrumentation and Measurement Technology Conference Proceedings of the 18th IEEE, 21-23 May 2001,): 2087-2092
    [83] S. B. CHEN, X. Z. CHEN. J. Q. LI and T. QIU. Acquisition of Weld Seam Dimensional Position Information for Arc Welding Robot Based on Vision Computing. Springer: Journal of Intelligent and Robotic Systems, 2005, 43, (1): 77-97
    [84] 谢杰成,张大力,徐文立.一种小波系数模型在图像噪卢参数估计中的应用.电子与信息学报,2004,26(5):673-678
    [85] 张永平,郑南宁,赵荣椿.基于变分的图像分割算法[J].中国科学(E辑),2002,32(1)133-144
    [86] Francis H Y, Chan F K Lam and Hui Zhu. Adaptive Thresholding by Variational Method. IEEE Transactions on Image Processing, 1998, 7: 468-473
    [87] Fei Liu, Xiaodan Song, Yupin Luo, Dongcheng Hu. Adaptive threshoiding based on variational background. LECTRONICS LETTERS, 2002, 38(18): 1017-1018
    [88] 曹利新,游洪,王晓明,刘健.面向截面测量数据B样条曲面光顺拟合及其精度计算[J].大连理工大学学报,2003,43(3):320-323
    [89] 蒋大为,李安平,蔡元虎.满足约束条件的光顺B样条曲线.西北工业大学学报,1996,14(3):463-466
    [90] 赵荣椿,迟耀斌,朱重光.图像分割技术进展.中国体视学与图像分析,1998,3(2):121-128
    [91] 魏弘博,吕振肃,蒋田仔,刘新艳.图像分割技术纵览.甘肃科学学报,2004,16卷(2):19-24
    [92]卜方玲,徐新.一种基于小波分析的SAR图像斑点噪声滤波算法.武汉大学学报(信息科学版),2001,26(4):315-319
    [93] Zhao Hua Yang, Zhao Bang Pu, Zhen Qiang Qi. Novel edge preserving multiscale filtering method based on mathematical morphology. International Conference on Machine Learning and Cybernetics, 2-5, Nov, 2003, Vol. 5: 2970-2975
    [94] Mi Hyun Kim, Kwang Hoon Sohn. Edge-preserving directional regularization technique for disparity estimation of stereoscopic images. ICCE, International Conference on Consumer Electronics, 22-24, June, 1999: 376-377
    [95] Heers J, Schnorr C, Stiehl H S, Globally convergent iterative numerical schemes for nonlinear variational image smoothing and segmentation on a multiprocessor machine. IEEE Transactions on Image Processing, 2001, 10(6): 852-864
    [96] Tasdizen T, Whitaker R. Anisotropic diffusion of surface normals for feature preserving surface reconstruction. Fourth International Conference on 3-D Digital Imaging and Modeling Proceedings, October, 2003: 353-360
    [97] Xin Wang, Wong B S, Chen Guan Tui. X-ray image segmentation based on genetic algorithm and maximum fuzzy entropy. IEEE Conference on Robotics, Automation and Mechatronics, 1-3 Dec, 2004, Vol.2: 991-995
    [98] He J et al. A comparison of binarization methods for historical archive documents. Eighth International Conference on Document Analysis and Recognition, 29 Aug.-1 Sept, 2005, Vol. 1: 538-542
    [99] J H Xue et al. A novel method for adaptive enhancement and unsupervised segmentation of MRI brain image. IEEE International Conference on Acoustics, Speech, and Signal Processing, 7-11, May, 2001, vol. 3: 2013-2016
    [100] Feigin M, Sochen N. Segmentation and denoising via an adaptive threshold Mumford-Shah-like functional. Proceedings of the 17th International Conference on Pattern Recognition, 23-26, A ug, 2004, Vol. 2: 98-101
    [101] Manay S, Yezzi A. Anti-geometric diffusion for adaptive thresholding and segmentation. International Conference on Image Processing, 7-10, Oct, 2001, Vol. 2: 829-832
    [102] 刘岩.红外图像分析方法与目标识别研究.西北工业大学博士学位论文,2004
    [103] 王立.视觉机制研究及其在红外成像制导中的应用.西北工业大学博士学位论文,2004
    [104] Schwartz E L. Spatial mapping in the primate sensory projection: analytic structure and relevance to perception. Biol. Cybernetics, 1977, 25: 181-194
    [105] 王延平,赵恒卓,蔡芹.刚二维阶次规正不变矩识别三维目标.武汉大学学报(自然科学版),1996,42(5):639-646
    [106] 董武,李树祥.矩特征的一种快速算法[J].中国图像图像学报.1999,4(10):860-864.
    [107] 吴盘龙.防区外空面导弹制导系统关键技术研究.西北工业大学博士学位论文,2005
    [108] Kourosh J K, Hamid S Z. Radon transform orientation estimation for rotation invariant texture analysis. IEEE transactions on pattern analysis and machine intelligence, 2006, 27(6): 1004-1008
    [109] 郦苏丹.SAR图像特征提取与目标识别方法研究.国防科学技术大学博士学位论文,2001
    [110] A Kadyrov, M Petrou. The trace transform and its application. PAMI, 2001, 23: 811-828
    [111] M Petrou, A Kadyrov. Features invatiant to affine distortions from the trace transform. ICIP 2001, Thessaloniki, Greece, October, 2001: 852-855
    [112] 孙即祥.图像分析.北京:科学出版社,2005
    [113] 周秀芝,刘方,王润生.利用局部不变特征识别复杂平面多边形.计算机辅助设计与图形学学报,2003,15(7):858-862
    [114] 张元,王广宇,吕运朋.数字图像的不变特性与特征提取.郑州大学学报(自然科学版),2000,32(4):57-59
    [115] Jan Flusser, Tomas Suk. Pattern recognition by affine moment invariants. Pattern Recognition, 1993, 26(1): 167-174
    [116] 李迎春,陈贺新,赵铭等.基于飞机目标的仿射不变矩研究.吉林大学学报(信息科学版),2003,21(5):84-88
    [117] 张洪钺,杨雪勤,郭红涛.基于仿射变换的飞机图像识别.航空学报,2003,24(3):252-254
    [118] Mahmoud I K, Mohamed M B. A dyadic wavelet affine invariant function for 2D shape recognition. IEEE Trans Pattern Analysis and Machine Intelligence, 2001, 23(10): 1152-1164
    [119] T H Reiss. The revised Fundamental theorem of moment invariants. IEEE Trans Pattern Analysis and Machine Intelligence, 1991, 13(8): 830-833
    [120] J Flusser, T Suk. Affine moment invariants: a new tool for character recognition. Pattern Recognition Letters, 1994, 15(14): 433-436
    [121] Dongming Zhao, Jie Chen. Affine curve moment invariants for shape recognition. Pattern Recognition, 1997, 30(6): 895-901
    [122] A E Oirrak, M Daoudi, D Aboutajdine. Affine invariant descriptors using fourier series. Pattern Recognition Letters, 2002, 23(10): 1109-1118.
    [123] M I Khalil, M M Bayoumi. Affine invariants for object recognition using the wavelet transform. Pattern Recognition Letters, 2002, 23(123): 57-72
    [124] Y Avtithis, Y Xirouhakis, S Kollias. Affine-invariant curve normalization for object shape representation, classification, and retrieval. Machine Vision and Application, 2001, (13): 80-94
    [125] D Shen, Horace H S. Generalized affine invariant image normalization. IEEE Trans Pattern Analysis and Machine Intelligence. 1997, 19(5): 431-439
    [126] 张奇峰,张立明.一种新的统计仿射不变量.计算机工程与科学,2004,26(9):35-38
    [127] 田谨,蒋强荣,许蔓苓.基于多视点不变量的三维物体识别.北京工业大学学报,2000,26(1):98-101
    [128] 潘乐义.郑链.王克勇.红外成像制导中的三维目标识别.红外技术,2000,22(6):1-6
    [129] Barrett E B, Brill M H, Haag N N, et al. Invariant linear methods in photogrammetry and model-matching. Geometric lnvariance in Computer Vision, Chapter 14, 1992: 277-292
    [130] 张海燕,宋克欧,王东木.基于方面图技术的三维运动目标识别.中国图象图形学报,2003,8(8):866-87 l
    [131] 栾新,朱铁一.多视点识别三维运动目标新方法.机器人,1999,21(4):241-248
    [132] 任海萍,平子良等.用雅可比傅里叶矩进行图像描述.光学学报,2004,24(1):5-10
    [133] Prokop R J, Reeves A P. A survey of moment-based techniques for unoccluded object representation and recognition. Graphical Models and Image Processing, 1992, 54(5): 438-460
    [134] 杜亚娟,张洪才,潘泉.基于矩特征的三维飞机目标识别.数据采集与处理,2000,15(3):390-394
    [135] 唐国良,荆仁杰,姚庆栋.基于矩特征的三维飞机目标识别.电子学报,1995,23(4):88-90
    [136] 曾接贤,陈震,储君,黎明.基于形态图表示的三维物体识别算法.计算机应用,2003,23(2):9-12
    [137] 彭辉,黄士科,陶琳,张天序.基于Zemike矩的三维目标多视点特性视图建模.红外与激光工程,2005,34(3):292-296
    [138] Ziv Gigus, Jitendra Malik. Computing the aspect graph for linedrawing of polyhedral objects, IEEE Transactions on Pattern Analysis and Machine Intelligence, 1996, 12(2): 113-122
    [139] 姜红军,海鹰,平子良.用圆谐-傅立叶矩描述图像.内蒙古师范大学学报自然科学(汉文) 版,2003,32(4):344-346
    [140] Teague M R. Image analysis via the general theory of mements. J. Opt. Soc. Am, 1980, (70): 920-930
    [141] 金敏,徐守时,汪行.不变矩在模式识别中的应用研究.计算机工程与应用,2004(25):65-67
    [142] 丁兴号,邓善熙.Hu矩和Zemike矩在字符识别中的应用.工具技术,2003,37(13):16-18
    [143] 姜璐,舒华忠,章品正.Hu矩和Zemike矩在表情识别应用中的比较.洛阳大学学报,2004,19(2):14-17
    [144] 苏开娜,乔晓梅.基于Zemike矩的人体行为识别.北京工业大学学报,2005,31(4):423-426
    [145] Sheng Y L, Shen L X. Orthogonal Fourier-Mellin moments for invariant pattern recognition. J. Opt. Soc. Am., 1994, 11(6): 1748-1757
    [146] 阿木古楞,杨性愉,平子良.用变形雅可比(p=4,q=3)-傅立叶矩进行图像描述.光电子·激光,2003,14(9):981-985
    [147] 阿木古楞,杨性愉,平子良.变形雅可比(p=4,q=3)-傅立叶矩的抗噪声能力研究.光电子·激光,2003,14(11):1191-1195
    [148] Ping Z L, Wu R G, Sheng Y L. Image describtion with Chebyshev-Fourier Moments. J. Opt. Soc. Am. A, 2002, 19: 1748-1754
    [149] 平子良,盛云龙.用切比雪夫图像矩描述图像.内蒙古师范大学学报(自然科学汉文版),2002,31(3):216-219
    [150] 王海波,平子良.几种正交矩描述图像的性能比较.内蒙古师范大学学报(自然科学汉文版),2005,34(1):35-39
    [151] 海鹰,姜红军,平子良.圆谐.傅立叶矩图像重建误差分析.内蒙古师范大学学报自然科学(汉文)版,2003,32(4):347-349
    [152] 范立南.基于多结构元形态边缘检测与不变矩特征模式识别研究.东北大学博士学位论文,2004
    [153] 王晓红.矩技术及其在图像处理与识别中的应刚研究.西北工业大学博士学位论文,2001
    [154] 胡广书.数字信号处理-理论、算法与实现.北京:清华大学出版社,1997
    [155] 舒宁,马洪超,孙和利.模式识别的理论与方法.武汉:武汉大学出版社,2005
    [156] 边肇祺,张学工.模式识别.北京:清华大学出版社,2000
    [157] 陈爱军.基于人脸信息识别的信息安全技术研究.东北林业大学硕士学位论文,2002
    [158] 赵海涛.投影分析在人脸识别中的研究与应用.南京理工大学博士学位论文,2003
    [159] 王家生,刘嘉锟.随机过程基础.天津:天津大学出版社,2003
    [160] 陈良均,朱庆棠.随机过程及应用.北京:高等教育出版社,2003
    [161] Clarke R J. Transform coding of image. London: Academic Press, 1985
    [162] Jain A K. A sinusoidal family of unitary transform. IEEE Trans. Pattern Anal. Math. Intell., 1979, 1(4): 356-365
    [163] Ziad M, Martin D. Face recognition using the discrete cosine transform. International Journal of Computer Vision, 2001, 43(3): 167-188
    [164] 张敏贵.基于小波和支持向量机的人脸识别方法研究.两北工业大学博士学位论文,2003
    [165] Vapnik V, Lerner A. Pattern recognition using generalized portrait. Automation and Remote Control, 1963, 24(6): 774-780
    [166] Vapnik V. Estimation of Dependencies Based on Empirical Data. Berlin: Springer-Verlag, 1982
    [167] Vapnik V. The nature of statistical learning theory. NewYork: Springer-Verlag, 1995
    [168] Vapnik V. Statistical learning theory. NewYork: Wiley, 1998
    [169] Boser B E, Guyon I M, Vapnik V. A training algorithm for optimal margin classifiers. Proceedings of the fifth annual workshop on computational learning Theory, Pittsburgh, PA, USA, 1992, 144-152
    [170] Cones C, Vapnik V. Support vector networks. Machine Learning, 1995, 20(3): 273-297
    [171] Vapnik V, E. Levin and C. Y. Le. Measuring the VC-dimension of a learning machine. Neural Computation, 1994, 6: 851-876
    [172] Vapnik V. An overview of statistical learning theory. IEEE Transactions on Neural Networks, 1999, 10(5): 988-999
    [173] 王红梅.多传感器图像融合及其预处理技术研究.西北工业大学博士学位论文,2005
    [174] 王亮申,朱玉才,陈少华等.利用SVM进行车型识别.计算机工程与设计,2005,26(9):2453-2454
    [175] 安金龙,王正欧,马振平.一种新的支持向量机多类分类方法.信息与控制,2004,33(3):262-267
    [176] 郑勇涛,刘玉树.支持向量机解决多分类问题研究.计算机上程与应用,2005,(23):190-192
    [177] 王立国,张晔,谷延锋.支持向量机多类目标分类器的结构简化研究.中国图象图形学报,2005,10(5):571-574
    [178] 唐发明,王仲东,陈绵云.支持向量机多类分类算法研究.控制与决策,2005,20(7):746-749
    [179] 王建芬,曹元大.支持向量机在大类别数分类中的应用.北京理工大学学报,2001,21(2):225-228
    [180] 祁亨年.支持向量机及其应用研究综述.计算机工程,2004,30(10):6-9
    [181] 刘志刚等.支持向量机在多类分类问题中的推广.计算机工程与应用,2004,(7):10-13
    [182] 安金龙.支持向量机若干问题的研究.天津大学博士学位论文,2004
    [183] 郭世杰.基于支持向量机的多类分类问题的研究.上海师范大学硕士学位论文,2005
    [184] 黄琼英.支持向量机多类分类算法的研究及应用.河北工业大学硕士学位论文,2005
    [185] J C Platt. Fast Training of support vector machines using sequential minimal optimization. In Scholkopf B. et al (ed.), Advances in Kernel Methods-Support Vector Learning, Cambridge, MA, MIT Press, 1999, 185-208
    [186] Bottou L, Cortes C, Denker J. Comparison of classifier methods: a ease study in handwriting digit recognition. Proceedings of the 12th IAPR International Conference on Pattern Recognition. Jerusalem: IEEE, 1994: 77-82
    [187] Krebel U. Pairwise classification and support vector machines. Advances in Kernel Methods-Support Vector Learning. Cambridge, MA: MIT Press, 1999: 255-268
    [188] Crammer K, Singer Y. On the learnability and design of output codes for multiclass problems. Proceedings of the 13th Annual Conference on Computational Learning Theory. San-Fransisco: Morgan Kaufmann, 2000: 35-46
    [189] Hsu C W, Lin C J. A comparison of methods for multiclass support vector machines. IEEE Transactions on Neural Networks, 2002, 13(2): 415-425
    [190] Platt J C, Cristianini N, Shawe T J. Large margin DAGs for multiclass classification. Advances in Neural Information Proeessing Systems, 2000: 547-553
    [191] 王宏漫,欧宗瑛.基于支持向量机的人脸识别方法研究.小型微型计算机系统,2004,25(1):139-142
    [192] 谷春亮,尹宝才,孔德慧等.基于三维多分辨率模型与Fisher线性判别的人脸识别方法.计算机学报,2005,28(1):97-103
    [193] 李映,焦李成.基于核Fisher判别分析的目标识别.西安电子科技大学学报(自然科学版),2003,30(2):179-182
    [194] 孔锐,张国宜,施泽生等.基于核Fisher判决分析的脸谱识别新方法.电路与系统学报,2003,8(5):57-61
    [195] 李应红等.支持向量机的工程应用.北京兵器工业出版社,2004
    [196] 李晓宇,张新峰,沈兰荪.支持向量机的研究进展.测控技术,2006,25(5):7-12
    [197] 王怀野,张科,李言俊.一种自适应各项异性高斯滤波方法.计算机工程与应用,2004,40(10):18-19

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700