遥感图像高精度并行监督分类技术研究
|
摘要
遥感作为近几十年迅速发展起来的一门综合性技术学科，因其具有观测范围广、采集信息量大、获取信息速度快等特点，已经在民用和军用的众多领域发挥了重大作用。遥感图像处理是遥感科学与具体应用相结合的重要技术途径。遥感图像分类是遥感图像处理的一个重要内容，其中监督分类作为一种先学习后分类的机器学习策略，是对遥感图像进行定量分析的主要手段，应用领域十分广阔。
随着传感器、遥感平台、数据通信等相关技术的发展，通过遥感手段获取的数据量急剧膨胀，迫切需要快速遥感图像处理技术的支持。同时，各应用领域对遥感图像的处理速度和分析结果的量化程度要求越来越高。高精度、快速的遥感图像监督分类技术是遥感科学走向实用化、产业化的一个重要突破口。
遥感图像监督分类处理包括分类预处理、分类判别和分类后处理三个阶段。分类判别阶段的学习算法是影响监督分类精度的关键因素。分类预处理阶段的几何校正和分类判别阶段的学习与分类计算复杂度高，是导致遥感图像监督分类处理速度慢的主要原因。本文针对上述应用需求展开深入研究，首先提出几种学习算法用于提高监督分类的预测精度；然后采用并行处理技术，提高遥感图像监督分类的处理速度；最后设计并实现了一个遥感图像并行处理系统。本文的主要创新如下：
1、提出一种通用性好的纠错输出编码方法——搜索编码法。
有效地将多类分类问题转化为多个两类问题，可以拓展众多机器学习算法在遥感领域中的应用。纠错输出编码不仅可将多类问题两类化，而且能够提高监督分类的精度。但是目前没有一种编码算法适用于含任意类别数的监督分类任务。为解决这个问题，本文提出了一种搜索编码法SCM，该算法将二进制位串与非负整数对应起来，通过对整数空间的顺序搜索，获得满足任意类别数和最小汉明距离要求的输出码。将SCM用到多种监督学习算法后，实验结果表明该编码方法不仅适用范围广，而且能够有效地提高预测精度。
2、提出基于搜索编码的结构化神经网络。
Serpico和Roli根据遥感领域中多传感器图像的分类问题，提出了具有特殊结构的神经网络。本文将这两种神经网络结构分别扩展为结构化神经网络SNN和组合式结构化神经网络k-SNN。SNN的网络结构容易确定、网络行为易于理解，但是预测精度较低。k-SNN虽然可以提高SNN的预测精度，但是学习时间过长。为此，本文进一步提出基于搜索编码的结构化神经网络CSNN，并提出一种解释结构化神经网络分类行为的方法。CSNN在保持可理解性的同时，大大提高了结构化神经网络的预测精度。
3、提出基于属性变换的动态离散化方法RCAT，设计并实现了二分决策树系统Btrees。
不同光谱波段的像元灰度值是遥感图像监督分类的基本数据源，属于连续属性。由于
国防科学技术大学研究生院学位论文
一些重要的监督学习算法只能处理离散属性，对于连续属性的处理需要将连续属性离散
化，因此离散化方法的优劣直接关系到监督分类的预测精度。为此，本文提出基于属性变
换的动态多区间离散化方法RCAT。RCAI，将待处理的连续属性转化为一个概率属性，通
过对概率属性的二分离散化处理，获得原连续属性的一个多区间划分。在RCAT算法的基
础上，设计并实现了二分决策树系统BtreeS。与其它离散化方法相比，采用RCAT算法的
BtreeS系统生成的决策树不仅预测精度高，而且可理解性强。
4、针对分布存储的并行环境，提出两种基于不规则区域计算的并行图像扭曲算法。
分类预处理阶段的几何校正属于数字图像处理领域中的大图像扭曲问题，计算复杂度
高、处理时间长，是影响遥感图像监督分类性能的一个关键因素。通过分析现有研究中相
关算法的缺陷，本文提出两种基于不规则区域计算的并行图像扭曲算法PIWA一LOC和
PI认叭一LIC。PIWA一LOC算法对输入图像进行规则划分，各计算结点通过局部输出区域计算
确定本地负载，从而获得很好的数据局部性。PI场rA一LIC算法对目标图像的计算负载进行
均衡分配，并通过局部输入区域计算，确定各计算结点所需的输入数据，算法不仅数据局
部性好，而且能够实现负载均衡。实验结果表明，对于遥感图像的几何校正问题，这两种
并行图像扭曲算法均能获得很好的并行性能，显著提高了分类预处理的速度。其中，
PIWA一LIC算法具有更好的通用性。
5、针对分布存储的并行环境，对几种监督学习算法进行并行化处理。
对于遥感图像的监督分类问题，一般训练样本和待分类样本的数目庞大，导致学习与
分类速度缓慢;另一方面，人们在追求高精度的分类结果的同时，不可避免地增加了学习
和分类的复杂度。为此，本文针对分布存储的并行环境，对最近邻法的分类过程、以及论
文中所研究的四种神经网络算法的学习过程进行并行化处理。实验结果表明，监督学习算
法的并行化能够有效地提高分类判别阶段的处理速度。
基于上述研究成果，结合我国遥感应用领域对高性能遥感图像处理软件的迫切需求，
设计并实现了一个遥感图像并行处理系统YH一PRIPS。YH一P租PS采用基于B/S结构的三层
应用模型，并运用高性能并行手段实现遥感图像处理。由于系统采用层次结构与模块?
As a synthetic subject, remote sensing has gotten rapid development in recent decades. It plays an important role in civil and military applications as it provides a unique perspective from which to observe large regions and acquire abundant useful information in high speed. Remote-sensing image processing is dedicated to processing and analyzing digital images acquired by remote sensors. As a primary quantitative means for image analysis, supervised remote-sensing image classification is a crucial issue during the whole procedure of image processing, and has comprehensive applications in many domains.With the development of remote sensors, remote platforms, and data communication technologies, more and more remote sensing data are produced for processing. At the same time, applications have an increasing demand for high processing speed and accuracy. Thus, fast processing technologies for supervised remote-sensing image classification with high accuracy become more and more urgent, and come to be the focus of research.The procedure of supervised remote-sensing image classification can be divided into three phases: preprocessing phase, learning and discrimination phase, postprocessing phase. The learning algorithm of the second phase is critical for the accuracy of classification. Fastening geometric correction in the first phase and supervised learning step in the second phase are two key issues to improve processing speed. This thesis aims at improving the classifying accuracy and processing speed for supervised remote-sensing image classification, and makes the following contributions:1. A general coding method SCM is presented, which can generate error-correcting output codes suitable for the classification problem with any number of classes.Reducing a multi-class problem to multiple two-class problems can extend the applications for many learning algorithms. Error-correcting output codes (ECOCs) approach has this function as well as improve classifying accuracy. However, there is no coding method that can generate ECOCs suitable for any number of classes. Thus, this thesis proposes a search coding method (SCM) that associates nonnegative integers with binary strings. Given any number of classes and an expected minimum hamming distance, SCM can find out a satisfying output code through searching an integer range. Applied to several supervised learning algorithms, SCM is demonstrated to improve the recognition accuracy for both stable and unstable classifiers efficiently.2. A structured neural network based on SCM (CSNN) is explored.Serpico and Roli have proposed two neural networks with special structures devoted to
multi-sensor image recognition. This thesis extends the neural networks to a structured neural network (SNN) and a combined structured neural network (k-SNN). SNN has deterministic network structure and interpretable network behavior. However, its classifying accuracy is not high enough. Though K-SNN can improve the recognition accuracy of SNN to some extent, its learning time increases significantly. Based on the search coding method SCM, a structured neural network (CSNN) is presented; meanwhile a simple method to interpret the behavior of the structured neural networks is proposed. Experimental results show that CSNN gets the best classifying accuracy among the three structured neural networks while retaining the intelligibility.3. A dynamic range-splitting method based on continuous attributes transform (RCAT) is explored. Based on RCAT, a binary-classification tree system (Btrees) is implemented.Gray values of multi-spectral images are main attributes for supervised remote-sensing image classification. Discretization of these continuous attributes has a great impact on the classification result for many supervised learning algorithms. This thesis explores a dynamic discretization method named RCAT. RCAT uses simple binarization to get a multi-splitting result through mapping a continuous attribute into a probability attribute, where the binarization of the probability attribute corresponds to a mu
随着传感器、遥感平台、数据通信等相关技术的发展，通过遥感手段获取的数据量急剧膨胀，迫切需要快速遥感图像处理技术的支持。同时，各应用领域对遥感图像的处理速度和分析结果的量化程度要求越来越高。高精度、快速的遥感图像监督分类技术是遥感科学走向实用化、产业化的一个重要突破口。
遥感图像监督分类处理包括分类预处理、分类判别和分类后处理三个阶段。分类判别阶段的学习算法是影响监督分类精度的关键因素。分类预处理阶段的几何校正和分类判别阶段的学习与分类计算复杂度高，是导致遥感图像监督分类处理速度慢的主要原因。本文针对上述应用需求展开深入研究，首先提出几种学习算法用于提高监督分类的预测精度；然后采用并行处理技术，提高遥感图像监督分类的处理速度；最后设计并实现了一个遥感图像并行处理系统。本文的主要创新如下：
1、提出一种通用性好的纠错输出编码方法——搜索编码法。
有效地将多类分类问题转化为多个两类问题，可以拓展众多机器学习算法在遥感领域中的应用。纠错输出编码不仅可将多类问题两类化，而且能够提高监督分类的精度。但是目前没有一种编码算法适用于含任意类别数的监督分类任务。为解决这个问题，本文提出了一种搜索编码法SCM，该算法将二进制位串与非负整数对应起来，通过对整数空间的顺序搜索，获得满足任意类别数和最小汉明距离要求的输出码。将SCM用到多种监督学习算法后，实验结果表明该编码方法不仅适用范围广，而且能够有效地提高预测精度。
2、提出基于搜索编码的结构化神经网络。
Serpico和Roli根据遥感领域中多传感器图像的分类问题，提出了具有特殊结构的神经网络。本文将这两种神经网络结构分别扩展为结构化神经网络SNN和组合式结构化神经网络k-SNN。SNN的网络结构容易确定、网络行为易于理解，但是预测精度较低。k-SNN虽然可以提高SNN的预测精度，但是学习时间过长。为此，本文进一步提出基于搜索编码的结构化神经网络CSNN，并提出一种解释结构化神经网络分类行为的方法。CSNN在保持可理解性的同时，大大提高了结构化神经网络的预测精度。
3、提出基于属性变换的动态离散化方法RCAT，设计并实现了二分决策树系统Btrees。
不同光谱波段的像元灰度值是遥感图像监督分类的基本数据源，属于连续属性。由于
国防科学技术大学研究生院学位论文
一些重要的监督学习算法只能处理离散属性，对于连续属性的处理需要将连续属性离散
化，因此离散化方法的优劣直接关系到监督分类的预测精度。为此，本文提出基于属性变
换的动态多区间离散化方法RCAT。RCAI，将待处理的连续属性转化为一个概率属性，通
过对概率属性的二分离散化处理，获得原连续属性的一个多区间划分。在RCAT算法的基
础上，设计并实现了二分决策树系统BtreeS。与其它离散化方法相比，采用RCAT算法的
BtreeS系统生成的决策树不仅预测精度高，而且可理解性强。
4、针对分布存储的并行环境，提出两种基于不规则区域计算的并行图像扭曲算法。
分类预处理阶段的几何校正属于数字图像处理领域中的大图像扭曲问题，计算复杂度
高、处理时间长，是影响遥感图像监督分类性能的一个关键因素。通过分析现有研究中相
关算法的缺陷，本文提出两种基于不规则区域计算的并行图像扭曲算法PIWA一LOC和
PI认叭一LIC。PIWA一LOC算法对输入图像进行规则划分，各计算结点通过局部输出区域计算
确定本地负载，从而获得很好的数据局部性。PI场rA一LIC算法对目标图像的计算负载进行
均衡分配，并通过局部输入区域计算，确定各计算结点所需的输入数据，算法不仅数据局
部性好，而且能够实现负载均衡。实验结果表明，对于遥感图像的几何校正问题，这两种
并行图像扭曲算法均能获得很好的并行性能，显著提高了分类预处理的速度。其中，
PIWA一LIC算法具有更好的通用性。
5、针对分布存储的并行环境，对几种监督学习算法进行并行化处理。
对于遥感图像的监督分类问题，一般训练样本和待分类样本的数目庞大，导致学习与
分类速度缓慢;另一方面，人们在追求高精度的分类结果的同时，不可避免地增加了学习
和分类的复杂度。为此，本文针对分布存储的并行环境，对最近邻法的分类过程、以及论
文中所研究的四种神经网络算法的学习过程进行并行化处理。实验结果表明，监督学习算
法的并行化能够有效地提高分类判别阶段的处理速度。
基于上述研究成果，结合我国遥感应用领域对高性能遥感图像处理软件的迫切需求，
设计并实现了一个遥感图像并行处理系统YH一PRIPS。YH一P租PS采用基于B/S结构的三层
应用模型，并运用高性能并行手段实现遥感图像处理。由于系统采用层次结构与模块?
As a synthetic subject, remote sensing has gotten rapid development in recent decades. It plays an important role in civil and military applications as it provides a unique perspective from which to observe large regions and acquire abundant useful information in high speed. Remote-sensing image processing is dedicated to processing and analyzing digital images acquired by remote sensors. As a primary quantitative means for image analysis, supervised remote-sensing image classification is a crucial issue during the whole procedure of image processing, and has comprehensive applications in many domains.With the development of remote sensors, remote platforms, and data communication technologies, more and more remote sensing data are produced for processing. At the same time, applications have an increasing demand for high processing speed and accuracy. Thus, fast processing technologies for supervised remote-sensing image classification with high accuracy become more and more urgent, and come to be the focus of research.The procedure of supervised remote-sensing image classification can be divided into three phases: preprocessing phase, learning and discrimination phase, postprocessing phase. The learning algorithm of the second phase is critical for the accuracy of classification. Fastening geometric correction in the first phase and supervised learning step in the second phase are two key issues to improve processing speed. This thesis aims at improving the classifying accuracy and processing speed for supervised remote-sensing image classification, and makes the following contributions:1. A general coding method SCM is presented, which can generate error-correcting output codes suitable for the classification problem with any number of classes.Reducing a multi-class problem to multiple two-class problems can extend the applications for many learning algorithms. Error-correcting output codes (ECOCs) approach has this function as well as improve classifying accuracy. However, there is no coding method that can generate ECOCs suitable for any number of classes. Thus, this thesis proposes a search coding method (SCM) that associates nonnegative integers with binary strings. Given any number of classes and an expected minimum hamming distance, SCM can find out a satisfying output code through searching an integer range. Applied to several supervised learning algorithms, SCM is demonstrated to improve the recognition accuracy for both stable and unstable classifiers efficiently.2. A structured neural network based on SCM (CSNN) is explored.Serpico and Roli have proposed two neural networks with special structures devoted to
multi-sensor image recognition. This thesis extends the neural networks to a structured neural network (SNN) and a combined structured neural network (k-SNN). SNN has deterministic network structure and interpretable network behavior. However, its classifying accuracy is not high enough. Though K-SNN can improve the recognition accuracy of SNN to some extent, its learning time increases significantly. Based on the search coding method SCM, a structured neural network (CSNN) is presented; meanwhile a simple method to interpret the behavior of the structured neural networks is proposed. Experimental results show that CSNN gets the best classifying accuracy among the three structured neural networks while retaining the intelligibility.3. A dynamic range-splitting method based on continuous attributes transform (RCAT) is explored. Based on RCAT, a binary-classification tree system (Btrees) is implemented.Gray values of multi-spectral images are main attributes for supervised remote-sensing image classification. Discretization of these continuous attributes has a great impact on the classification result for many supervised learning algorithms. This thesis explores a dynamic discretization method named RCAT. RCAT uses simple binarization to get a multi-splitting result through mapping a continuous attribute into a probability attribute, where the binarization of the probability attribute corresponds to a mu
引文
[1] Mather P.M. Computer Processing of Remotely-Sensed Images: An Introduction, (Second Edition). Chichester: John Wiley & Sons. 1999
[2] 张永生，王仁礼．遥感动态监测．北京：解放军出版社．1999
[3] 钱曾波，刘静宇，肖国超．航天摄影测量．北京：解放军出版社．1992
[4] 星上信息处理系统研究报告．[技术报告] ．国防科技大学计算机学院．2002．12
[5] 空间对抗技术的概念研究．[技术报告] ．国防科技大学计算机学院．2004．3
[6] 朱述龙，张占睦．遥感图像获取与分析．北京：科学出版社．2000
[7] 章孝灿，黄智才，赵元洪．遥感数字图像处理．杭州：浙江大学出版社．1997
[8] 张永生．遥感图像信息系统．北京：科学出版社．2000
[9] 余旭初．模式识别与图像分类．北京：解放军出版社．2000
[10] 熊桢．高光谱遥感图像分类技术研究．[博士论文] ．中国科学院遥感应用研究所．北京，2000
[11] Mitchell T.M. Machine Learning. McGraw-Hill Science/Engineering/Math, 1997
[12] 章毓晋．图像工程—图像处理和分析．北京：清华大学出版社．1999
[13] Lillesand T.M., Kiefer R.W., Chipman J.W. Remote Sensing and Image Interpretation. John Wiley&Sons, 2003
[14] 陈述彭．航天遥感应用的世纪畅想．计算机世界报，2004．1
[15] 骆剑承．遥感地学智能图解模型研究及其应用．[博士论文] ．中国科学院地理研究所．北京，1999
[16] WTO与中国航天遥感发展战略．中国航天，2002．7 http://www.space.cetin.net.cn/docs/ht0207/ht020707.htm
[17] Berger J．O．著，贾乃光译．统计决策论及贝叶斯分析．中国统计出版社．1998
[18] Cortes C., Vapnik V. Support-vector networks. Machine Learning. 1995, 20(3): 273-297
[19] Wolberg G. Digital Image Warping. IEEE Computer Society Press, Los Alamitos, CA, 1990
[20] Wolberg G, Sueyllam H.M., Ismail M.A., Ahmed K.M. One-dimensional resampling with inverse and forward mapping functions. Journal of Graphics Tools, 2001, 5(3): 11-33
[21] 李晓梅，莫则尧，胡庆丰等．可扩展并行算法的设计与分析．北京：国防工业出版社．2000
[22] 郑纬民，汤志忠．计算机系统结构(第二版)．北京：清华大学出版社．1998
[23] 曾泳泓，成礼智，周敏．数字信号处理的并行算法．长沙：国防科技大学出版社． 1999
[24] Pitas Ⅰ. Parallel Algorithms for Digital Image Processing, Computer Vision and Neural Network. John Wiley & Son Ltd. UK. 1993
[25] 杨勃．并行图像分割技术的研究．[博士论文] ．合肥：中国科学技术大学．1998
[26] 黄波，王英杰．GIS与ES的结合及其应用初探．环境遥感，1996，11(3)：234-239
[27] 术洪磊．基于知识的遥感影像分类与制图综合方法研究．[博士论文] ．北京：北京大学，1995
[28] 术洪磊，毛赞猷．GIS辅助下的基于知识的遥感影像分类方法研究．测绘学报，1997，4：329-336
[29] Zhang L.J., Liu X.W. The studying of expert system for soil remote sensing classification and recognition. In Proceedings of Geoscience and Remote Sensing Symposium (IGARSS'93), 1993, 1:186-190
[30] 游先祥．应用遥感复合方法的森林分类和动态监测研究．环境遥感，1995，9(2)：197-105
[31] Mesev V. The use of census data in urban image classification. Photogrammetric Engineering and Remote Sensing, May 1998, 64(5): 431-438
[32] 蒋艳凰，周海芳，杨学军．监督学习的发展动态．计算机科学．2003，30(7)：7-11
[33] Breiman L. Bagging predictors. Machine Learning. 1996, 24(2): 123-140
[34] Freund Y., Schapire R. E. Experiments with a new boosting algorithm. In Saitta, L.(Ed.), Proceedings of the Thirteenth International Conference on Machine Learning, San Francisco, CA. Morgan Kaufman. 1996: 148-156
[35] Parmanto B., Munro P. W., Doyle H. R. Improving committee diagnosis with resampling techniques. Advances in N.eural Information Processing System. Cambridge, MA. MIT Press. 1996, 8:882-888
[36] Cherkauer K. J. Human expert-level performance on a scientific image analysis task by a system using combined artificial neural networks. In Chan, Working Notes of the AAAI Workshop on Intergrating Multiple Learned Models, 1996:15-21
[37] Tumer K., Ghosh J. Error correlation and error reduction in ensemble classifiers. Connection Science. 1996, 8(3-4): 385-404
[38] Dietterich T., Bakiri G. Solving multiclass learning problems via error-correcting output codes. Journal of Artificial Intelligence Research, 1995, 2:263-286
[39] Kolen J. F., Pollack J. B. Back propagation is sensitive to initial conditions. In Advances in Neural Information Processing System. San Francisco, CA. Morgan Kaufrnann. 1991, 3: 860-867,
[40] Optiz D.W., Shavlik J.W. Generating accurate and diverse members of a neural-network ensemble. In Advances in Neural Information Processing System. Cambridge, MA. MIT Press. 1996, 8:535-541
[41] Buntine W.L. A theory of learning classification rules. Ph.D. thesis, University of Technology, School of Computing Science, Sydney, Australia, 1990
[42] Wolpert D. Stacked generalization. Neural Network, 1992, 5(2): 241-260
[43] Gama J. Combining classification algorithm. Ph.D. Thesis, 1999
[44] Giacinto G., Roli F. Dynamic classifier selection. Multiple Classifier Systems. 2000: 177-189
[45] Eadie D., Shevlin R, Nisbet A. Correction of Geometric Image Distortion Using FPGAs. Optical Metrology, Imaging, and Machine Vision Conference, Galway, IRL, SPIE-Int'l Society for Optical Engineering, 2002
[46] Flavio, L.C. Padua, Guilherme A.S. Permeira, etc. Improving Processing Time of Large Images by Instruction Level Parallelism. Workshop on Parallel and Distributed Systems, Punta Arenas, Chile, 2001
[47] Catmull E., Smith A.R. 3-D transformations of images in scanline order. Computer Graphics, (SIGGRAPH'80 Proceedings), July 1980, 14(3): 279-285
[48] Wolberg G, Boult T.E. Separable image warping with spatial lookup tables. ACM Computer Graphics, 1989, 23(3): 369-378
[49] Wolberg G., Massalin H.. Fast convolution with packed lookup tables. Graphics Gems IV, Ed. By P. Heckbert, Academic Press, 1994: 447-464
[50] Fant K.M. A nonaliasing, real-Time spatial transform technique. IEEE Computer Graphics and Applications, 1986, 6(1): 71-80
[51] Meinds K., Barenbrug B. Resample hardware for 3D graphics. SIGGRAPH/Eurographics Graphics Hardware workshop, 2002: 17-26
[52] Warpenburg M. R., Siegel L.J. SIMD Image Resampling. IEEE Transactions on Computers, 1982, 31(10): 934-942,
[53] Jenq J.F., Sahni S. Reconfigurable mesh algorithms for image shrinking, expanding, clustering and template matching. In Proceedings of the Fifth International Parallel Processing Symposium, 1991: 648-656
[54] Wittenbrink CM., Somani A.K. 2D and 3D optimal image warping. Journal of Parallel and Distributed Computing. 1995, 25(2): 197-208
[55] Sylvain C.V., Serge M. A load-balanced algorithm for parallel digital image warping. International journal of pattern recognition and artificial intelligence, 1999, 13(4): 445-463
[56] Musick R., Catlett J., Russell S. Decision theoretic subsampling for induction on large database. In Utgoff, Proceedings of the Tenth International Conference on Machine Learning. San Francisco, CA. Morgan Kaufmann. 1992: 212-219
[57] Shafer J., Agrawal R., Mehta M. SPRINT: A scalable parallel classifier for data mining. In proceedings of the Twenty-second VLDB Conference. San Francisco, CA. Morgan Kaufmann. 1996:544-555
[5
[58] Ruggieri S. Efficient C4.5. IEEE Transactions on Knowledge and Data Engineering, 2001, 14(2): 438-444
[59] Chan RK., Stolfo S.J. Learning arbiter and combiner trees from partitioned data tbr scaling machine learning. In Proceedings of the First International Conference on Knowledge Discovery and Data Mining. Menlo Park, CA. AAAI Press. 1995:39-44
[60] Bowyer K., Chawla N., Moore T., Hall L., Kegelmeyer W. A parallel decision tree builder for mining very large visualization datasets. In Proceedings of IEEE-System, Man and Cybernetics, 2000
[61] Zaki M.J., Ho C.T, Agrawal R. Parallel Classification for Data Mining on Shared-Memory Multiprocessors. IBM Technical Report, 1998:198-205
[62] Joshi M.V., Karypis G, Kumar V. ScalParC: A New Scalable and Efficient Parallel Classification Algorithm for Mining Large Datasets. In Proceedings of the International Parallel Processing Symposium, 1998
[63] Caccetta RA., Campbell N.A., West G.A. A massively parallel implementation of an image classifier. In Proceedings of the 16th Asian Conference on Remote Sensing, 1995
[64] 刘勇卫，贺学鸿．遥感精解．测绘出版社，1993年12月第一版，170-220
[65] Wharton, S.W. A spectral-knowledge-based approach for urban land-cover discrimination, IEEE Transactions on Geoscience and Remote Sensing, May 1987, GE-25(3): 272-282
[66] Serpico S.B., Roli F. Classification of multisensor remote-sensing images by structured neural networks. IEEE Transactions on Geoscience and Remote Sensing, 1995, 33(3): 562-578
[67] Roli F. Multisensor image recognition by neural networks with understandable behavior. International journal of pattern recognition and artificial intelligence, 1996, 10(8): 887-917
[68] Setiono R. Extracting Rules from Neural Networks by Pruning and Hidden-Unit Splitting. Neural Computation, 1997, 9(1): 205-225
[69] Craven M.W. Extracting comprehensible models from trained neural network. Ph. D thesis, University of wisconsin, madison, 1996
[70] Bay, S.D. UCI KDD Archive [http://kdd.ics.uci.edu] , 1999
[71] Jiang Y.H., Zhao Q.L., Yang X.J. A general coding method for error-correcting output codes. LNCS 3056/PAKDD2004, Sydney Australia, May 26-28, 2004, 已录用
[72] 蒋艳凰，杨学军．基于搜索输出编码的简单贝叶斯分类方法．国防科学技术大学学报，已录用
[73] Quinlan, J.R. C4.5: Programs for Machine Learning. San Mateo, CA: Morgan Kaufmann Publishers, Inc., 1993
[7
[74] Breiman L., Friedman J.H., Olshen R.A., Stone C.J. Classification and Regression Trees. Wadsworth Ⅰ nternational Group, 1984
[75] Rumelhart D.E., Hinton G., Williams R. Learning Internal Representations through Error Propagation. In Rumelhart D.E., McClelland J.L. & the PDP research Group, (Eds.), Parallel Distributed Processing: Experiments in the Microstructure of Cognition, Vol. 1. Cambridge: MIT Press, 1986
[76] Freund Y., Schapire R.E. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Science. 1997, 55(1): 119-139
[77] Amari S., Wu S. Improving support vector machine classifiers by modifying kernel functions. Neural Networks, 1999, 12(6): 783-789
[78] Allwein, R. E. Schapire, Singer Y. Reducing multiclass to binary: A unifying approach for margin classifier. Journal of Machine Learning Research, 2000, 1:113-141
[79] Hastie T., Tibshirani R. Classification by pairwise coupling. The Annals of Statistics, 1998, 26(2): 451-471
[80] Sejnowski T.J., Rosenberg C.R. Parallel networks that learn to pronounce English text. Journal of complex systems, 1987, 1(1): 145-168
[81] 肖国镇，卿斯汉．编码理论．北京：国防工业出版社，1993
[82] Bose R.C, Ray-Chaudhuri D.K. On a class of error-correcting binary group codes. Information and Control, 1960, 3(1): 68-79
[83] Crammer K., Singer Y. On the learnability and design of output codes for multiclass problems. In Proceedings of the Thirteenth Annual 'Conference on Computational Learning Theory, 2000:35-46
[84] Dietterich T. Machine learning research: four current directions. AI Magazine, 1997, 18(4): 97-136
[85] Hansen L., Salamon P. Neural network ensembles. IEEE Trans. Pattern Analysis and Machine Intell. 1990, 12: 993-1001
[86] 蒋艳凰，杨学军．多层组合分类器研究．计算机工程与科学，已录用
[87] Kong E.B, Dietterich T.G. Error-Correcting Output Coding Corrects Bias and Variance. In proceedings of the 12th International Conference on Machine Learning. In S. Prieditis and S. Russell, eds. 1995:313-321
[88] Gama J. Combining classification algorithm. Ph.D. Thesis, 1999
[89] Kohavi R., Wolpert D.H. Bias plus variance decomposition for zero-one loss functions. In Proceedings of the 13th International Conference on Machine Learning. Morgan Kaufmann, 1996:275-283
[90] Langley P. Induction of recursive Bayesian classifiers. In Proceedings of the 6th European Conference on Machine Learning. Lecture Notes in Artificial Intelligence, 1993, 667:153-164
[9
[91] 周志华，陈世福．神经网络集成．计算机学报，2002，25(1)：1-8
[92] Ting, K.M., Zheng, Z.J. Improving the performance of boosting for naive Bayesian classification. In Proceedings of Pacific-Asia Conference on Knowledge Discovery and Data Mining, Berlin: Springer-Verlag. 1999:296-305
[93] 吴翊，李永乐，胡庆军．应用数理统计．国防科技大学出版社，1997
[94] Cestnik B. Estimating probabilities: a crucial task in machine learning. In proceedings of ECAI 90, Stockholm, 1990
[95] Dzeroski S., Cestnik B., Petrovski I. Using the m-estimate in rule induction. Journal of Computing and Information Technology, 1993, 1(1): 37-46
[96] Cybenko G. Continuos valued neural networks with two hidden layers are sufficient. Technical Report, Department of Computer Science, Tufts University, Medford, MA, 1988
[97] Jimenez D. Dynamically weighted ensemble neural networks for classification. In proceedings of IJCNN98, Vol. 1. Anchorage, AK, IEEE Computer Society Press, Los Alamitos, CA, 1998:753-756
[98] Krogh A., Vedelsby J. Neural network ensembles, cross validation, and active learning. In Tesauro G., Touretzky D., Leen T. (Eds.), Advances in Neural Information Processing Systems 7, Denver, CO, MIT Press, Cambridge, MA, 1995:231-238
[99] Lippman R.P. An introduction to computing with neural nets. IEEE ASSP Magazine, 1987, 2:4-22
[100] Kanellopoulos I., Wilkinson G.G. Strategies and best practice for neural network image classification. International Journal of Remote Sensing, 1997, 18(4): 711-725
[101] Hwang J.N., Lay S.R., Kiang R. Robust construction neural networks for classification of remotely sensed data. In Proceedings World Congress on Neural Network. Portland, July 11-15, 1993, 4:580-584
[102] 蒋艳凰，周海芳，杨学军．一种结构化神经网络及其特性分析．计算机研究与发展．2002，39(增刊)：291-297
[103] 蒋艳凰，周海芳，杨学军．基于纠错编码的CSNN及其在遥感图像分类中的应用．计算机研究与发展，2003，40(7)：918-924
[104] Hertz J., Krogh A., Palmer R.G. Introduction to the Theory of Neural Computation. Reading, MA: Addison Wesley. 1991
[105] Peng C.H., Wen X.Z. Recent Applications of Artificial Neural Networks in Forest Resource Management: An Overview. Environmental Decision Support Systems and Artificial Intelligence. Menlo Park, CA: AAAI press, 1999:15-22
[106] 袁曾任．人工神经元网络及其应用．北京：清华大学出版社．1999
[107] Shavlik J.W., Mooney R.J., Towell G.G. Symbolic and neural learning algorithms: an experimental comparison. Machine Learning. 1991, 6(1): 111-143
[108] Chandrasekaran B., Goel A. From numbers to symbols to knowledge structures: Artificial intelligence perspective on the classification task. IEEE Trans, Syst, Man Cybernet. 1988, 18(3): 415-424
[109] Drago G.P, Ridella S. Statistically controlled activation weight initialization (SCAWI). IEEE Trans Neural Networks. July 1992, 3:627-631
[110] Darken C, Moody J. Note on learning rate schedules for stochastic optimization. Neural Information Processing Systems. Lippmann R.P, Moody J. E., Touretzky, D. S. (Editors) 1991:832-838
[111] Salomon R. Improved convergence rate of back-propagation with dynamic adaptation of the learning rate. Lecture Notes in Computer Science, RPSN Ⅰ. Dortmund, 1990:269-273
[112] Tollenaere T. SuperSAB: fast adaptive backpropagation with good scaling properties. Neural Networks. 1990, 3:561-573
[113] Jacobs R.A. Increased rates of convergence through learning rate adaptation. Neural Networks, 1988, 1(4): 295-308
[114] Riedmiller M., Heinrich B. A direct adaptive method for faster backpropagation learning: the RPROP algorithm. In Proceedings of the IEEE Intemational Conference on Neural Networks (ICNN). San Francisco, 1993:586-591
[115] Riedmiller M. Rprop-description and implementation details. Technical Report. Institut fur Logic Konplexitat and Deduktionssyteme University of Karlsruhe FRG, January 1994
[116] Igel C., Husken M. Improving the Rprop learning algodthrn. In Proceedings of the Second International Symposium on Neural Computation, NC'2000, ICSC Academic Press, 2000:115-121
[117] Simon H.A. The Science of the Artificial. Cambridge, MA: MIT. Press, 2nd edition, 1981
[118] 张朝晖，陆玉昌，张钹．发掘多值属性的关联规则．软件学报，1998，9(11)：801-805
[119] 张朝晖．数据采掘的多策略机器学习方法．[博士论文] ．清华大学计算机系，北京，1998
[120] Daugherty J., Kohavi R., Sahami M. Supervised and unsupervised discretization of continuous feature. In Proceedings of Twelfth International Conference on Machine Learning, San Francisco: Morgan Kaufrnann, 1995:194-202
[121] Jiang Y.H., Zhou H.E, Yang X.J. Constructing Decision Tree with Continuous Attributes for Binary Classification. IEEE Conference ICMLC2002, Beijng, China, November 2002, 4(2): 617-622
[122] Jiang Y.H., Yang X.J., Zhao Q.L. Constructing Decision Tree with High Intelligibility. Journal of Software. 2003, 14(12): 1996-2005
[123] Liu H., Hussain R, Tan C.L., Dash M. Discretization: an enabling technique. Data Mining and Knowledge Discovery, 2002, 6(4): 393-423
[124] Zemke S., Rams M. Multivariate Feature Coupling and Discretization. FEA-2003, US 2003
[125] Bay S. D. Multivariate discretization for set mining. Knowledge and Information Systems, 2001, 4(3): 491-512
[126] Holte R.C. Very simple classification rules perform well on most commonly used datasets. Machine Learning, 1993, 11: 63-90
[127] Quinlan J.R. Induction of decision trees. Machine Learning, 1986, 1: 81-106
[128] Quinlan J.R. Improved use of continuous attributes in C4.5. Journal of Intelligence Research, 1996,4:77-90
[129] Catlett J. On changing continuous attributes into ordered discrete attributes. Proceeding of the Fifth European Working Session on Learning, Heidelberg: Springer-Verlag, 1991: 164-178
[130] Fayyad U.M., Irani K.B. Multi-interval discretization of continuous-valued attributes for classification learning. Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence. San Francisco, CA: Morgan Kaufman, 1993: 1022-1027
[131] Fayyad U.M., Irani K.B. Discretizing continuous attributes while learning bayesian networks. In Proc. Thirteenth International Conference on Machine Learning. Morgan Kaufmann, 1996: 157-165
[132] Cerquides J., Mantaras R.L. Proposal and empirical comparison of a parallelizable distance-based discretization method. In KDD97: 3rd International Conference on Knowledge Discovery and Data Mining, 1997: 139-142
[133] Mantaras R.L. A distance based attribute selection measure for decision tree induction. Machine Learning, 1991: 103-115
[134] Ho K.M., Scott P.D. Zeta: A global method for discretization of continuous variables. In KDD97: 3rd International Conference on Knowledge Discovery and Data Mining, 1997: 191-194
[135] Kerber R. ChiMerge: Discretization of numeric attributes. In Proc AAAI-92, Ninth National Conference on Artificial Intelligence, pages 123-128. AAAI Press/The MIT Press, 1992
[136] Liu H., Setiono R. Feature selection and discretization. IEEE Transactions on Knowledge and Data Engineering, 1997, 9:1-4
[137] Wallace C.S., Patrick J.D. Coding decision trees. Machine Learning, 1993, 11:7-22
[138] Auer P., Holte R.C, Maass W. Theory and application of agnostic PAC-learning with small decision trees. In: Prieditis, A., Russell, S. Eds. Proceedings of the 12th International Conference on Machine Learning. San Francisco: Morgan Kaufmann Publishers, Inc., 1995: 21-29
[1
[139] Elomaa T., Rousu J. General and Efficient Multisplitting of Numerical Attributes. Machine Learning, 1999, 36:1-49
[140] Rousu J. Efficient range partitioning in classification learning. PhD thesis, Series of publication A, Report A-2001-1. University of Helsinki, 2001
[141] 史忠植．知识发现．北京：清华大学出版社．2002
[142] Quinlan J.R. Simplifying decision tree. Intemational Journal of Man-Machine Studies, 1987, 27:221-234
[143] Kenneth R．Castleman著．数字图像处理(Digital image processing)．北京：清华大学出版社．2003
[144] Buyya R. High Performance Cluster Computing: Architectures and Systems, volume 1. Prentice Hall PTR, Upper Saddle River, NJ, 1999
[145] 林凡．集群的可扩展性及其分布式体系结构(上和下)．IBM developerWorks中国网站．
http://www-900.ibm.com/developerWorks/cn/linux/cluster/clustersystem/base/part1/index.shtml
http://www-900.ibm.com/developerWorks/cn/linux/cluster/cluster_System/base/part2/indexl.shtml
http://www-900.ibm.com/developerWorks/cn/linux/cluster/cluster_system/base/part2/index2.shtml
[146] Jiang Y.H., Yang X.J., Dai H.D., Yi H.Z. A Distributed Parallel Resampling Algorithm for Large Images. LNCS 2834/APPT2003, Xiamen, China, September 17-19, 2003:608-618
[147] 蒋艳凰，杨学军，易会战．基于Cluster环境的大图像并行重采样技术研究．计算机研究与发展，已录用
[148] 蒋艳凰，杨学军，易会战．卫星遥感图像并行几何校正算法研究．计算机学报，已录用
[149] Paul S. Heckbert. Survey of texture mapping. IEEE Computer Graphics and Application, November 1986:207-212
[150] Beier T., Neely S. Feature-based image metamorphosis. Computer Graphics, 1992, 26(2): 35-42.
[151] Chen S.E. Quicktime VR - an image-based approach to virtual environment navigation. Computer Graphics (SIGGRAPH 95), August 1995: 29-38
[152] Thévenaz P., Blu T., Unser M. Image interpolation and resampling. Handbook of Medical Imaging, Processing and Analysis, I.N. Bankman, Ed., Academic Press, San Diego CA, USA, 2000, 393-420
[153] Breene L.A., Bryant J. Image warping by scanline operations. Computers & Graphics, 1993, 17(2): 127-130
[154] Heckbert RS. Fundamentals of texture mapping and image warping. Technical Report, CS Division, EECS Department, UC Berkeley, May 1989
[155] Li Z-C. Analysis of discrete techniques for image transformation. Numerical Algorithms, 1996, 13(3-4): 225-263
[156] Zhang Y. A fuzzy approach to digital image warping. IEEE Computer Graphics & Applications, July 1996, 16(4): 34-41
[157] Foley J.D., Dam A.V., Feiner S.K., Hughes J.F. Computer Graphics, Principles and Practice, 2nd edition, Addison-Wesley, Reading, Massachusetts, 1990
[158] Keys R.G. Cubic convolution interpolation for digital image processing. IEEE transactions on Acoustics, Speech, and Signal Processing, 1981, 29(6): 1153-1160
[159] Lee C., Hamdi M. Parallel image processing applications on a network of workstations. Parallel Computing, 1995, 21:137-160
[160] 周海芳．遥感图像并行处理算法的研究与应用(博士论文)．长沙：国防科学技术大学，2003．10
[161] Hillis W.D., Steele G.L., Jr. Data parallel algorithms. Communcations of ACM (Special issue on parallism), 1986, 29(12): 1170-1183
[162] Serge M. and Yves R. Elastic load balancing for image processing algorithms. In HP Zima, editor, Parallel Computation, LNCS 591, Springer Verlag, 1992:438-451
[163] PCI Geomatics主页. http://www.pcigeomatics.com/
[164] ERDAS公司主页. http://www.erdas.com/ERDASredirect/index2.html
[165] RSI公司主页. http://www.rsinc.com/envi/index.cfm
[166] RSIM主页. http://159.226.117.45/rsim/index.htm
[167] Thain D., Livny M. The ethernet approach to grid computing. In the Proceedings of the Twelfth IEEE Symposium on High Performance Distributed Computing, Seattle, WA, 2003
[168] Nicholas M. Short, Sr. Remote sensing and image interpretation & analysis, http://mercator.upc.es/tutorial/table.html.
[2] 张永生，王仁礼．遥感动态监测．北京：解放军出版社．1999
[3] 钱曾波，刘静宇，肖国超．航天摄影测量．北京：解放军出版社．1992
[4] 星上信息处理系统研究报告．[技术报告] ．国防科技大学计算机学院．2002．12
[5] 空间对抗技术的概念研究．[技术报告] ．国防科技大学计算机学院．2004．3
[6] 朱述龙，张占睦．遥感图像获取与分析．北京：科学出版社．2000
[7] 章孝灿，黄智才，赵元洪．遥感数字图像处理．杭州：浙江大学出版社．1997
[8] 张永生．遥感图像信息系统．北京：科学出版社．2000
[9] 余旭初．模式识别与图像分类．北京：解放军出版社．2000
[10] 熊桢．高光谱遥感图像分类技术研究．[博士论文] ．中国科学院遥感应用研究所．北京，2000
[11] Mitchell T.M. Machine Learning. McGraw-Hill Science/Engineering/Math, 1997
[12] 章毓晋．图像工程—图像处理和分析．北京：清华大学出版社．1999
[13] Lillesand T.M., Kiefer R.W., Chipman J.W. Remote Sensing and Image Interpretation. John Wiley&Sons, 2003
[14] 陈述彭．航天遥感应用的世纪畅想．计算机世界报，2004．1
[15] 骆剑承．遥感地学智能图解模型研究及其应用．[博士论文] ．中国科学院地理研究所．北京，1999
[16] WTO与中国航天遥感发展战略．中国航天，2002．7 http://www.space.cetin.net.cn/docs/ht0207/ht020707.htm
[17] Berger J．O．著，贾乃光译．统计决策论及贝叶斯分析．中国统计出版社．1998
[18] Cortes C., Vapnik V. Support-vector networks. Machine Learning. 1995, 20(3): 273-297
[19] Wolberg G. Digital Image Warping. IEEE Computer Society Press, Los Alamitos, CA, 1990
[20] Wolberg G, Sueyllam H.M., Ismail M.A., Ahmed K.M. One-dimensional resampling with inverse and forward mapping functions. Journal of Graphics Tools, 2001, 5(3): 11-33
[21] 李晓梅，莫则尧，胡庆丰等．可扩展并行算法的设计与分析．北京：国防工业出版社．2000
[22] 郑纬民，汤志忠．计算机系统结构(第二版)．北京：清华大学出版社．1998
[23] 曾泳泓，成礼智，周敏．数字信号处理的并行算法．长沙：国防科技大学出版社． 1999
[24] Pitas Ⅰ. Parallel Algorithms for Digital Image Processing, Computer Vision and Neural Network. John Wiley & Son Ltd. UK. 1993
[25] 杨勃．并行图像分割技术的研究．[博士论文] ．合肥：中国科学技术大学．1998
[26] 黄波，王英杰．GIS与ES的结合及其应用初探．环境遥感，1996，11(3)：234-239
[27] 术洪磊．基于知识的遥感影像分类与制图综合方法研究．[博士论文] ．北京：北京大学，1995
[28] 术洪磊，毛赞猷．GIS辅助下的基于知识的遥感影像分类方法研究．测绘学报，1997，4：329-336
[29] Zhang L.J., Liu X.W. The studying of expert system for soil remote sensing classification and recognition. In Proceedings of Geoscience and Remote Sensing Symposium (IGARSS'93), 1993, 1:186-190
[30] 游先祥．应用遥感复合方法的森林分类和动态监测研究．环境遥感，1995，9(2)：197-105
[31] Mesev V. The use of census data in urban image classification. Photogrammetric Engineering and Remote Sensing, May 1998, 64(5): 431-438
[32] 蒋艳凰，周海芳，杨学军．监督学习的发展动态．计算机科学．2003，30(7)：7-11
[33] Breiman L. Bagging predictors. Machine Learning. 1996, 24(2): 123-140
[34] Freund Y., Schapire R. E. Experiments with a new boosting algorithm. In Saitta, L.(Ed.), Proceedings of the Thirteenth International Conference on Machine Learning, San Francisco, CA. Morgan Kaufman. 1996: 148-156
[35] Parmanto B., Munro P. W., Doyle H. R. Improving committee diagnosis with resampling techniques. Advances in N.eural Information Processing System. Cambridge, MA. MIT Press. 1996, 8:882-888
[36] Cherkauer K. J. Human expert-level performance on a scientific image analysis task by a system using combined artificial neural networks. In Chan, Working Notes of the AAAI Workshop on Intergrating Multiple Learned Models, 1996:15-21
[37] Tumer K., Ghosh J. Error correlation and error reduction in ensemble classifiers. Connection Science. 1996, 8(3-4): 385-404
[38] Dietterich T., Bakiri G. Solving multiclass learning problems via error-correcting output codes. Journal of Artificial Intelligence Research, 1995, 2:263-286
[39] Kolen J. F., Pollack J. B. Back propagation is sensitive to initial conditions. In Advances in Neural Information Processing System. San Francisco, CA. Morgan Kaufrnann. 1991, 3: 860-867,
[40] Optiz D.W., Shavlik J.W. Generating accurate and diverse members of a neural-network ensemble. In Advances in Neural Information Processing System. Cambridge, MA. MIT Press. 1996, 8:535-541
[41] Buntine W.L. A theory of learning classification rules. Ph.D. thesis, University of Technology, School of Computing Science, Sydney, Australia, 1990
[42] Wolpert D. Stacked generalization. Neural Network, 1992, 5(2): 241-260
[43] Gama J. Combining classification algorithm. Ph.D. Thesis, 1999
[44] Giacinto G., Roli F. Dynamic classifier selection. Multiple Classifier Systems. 2000: 177-189
[45] Eadie D., Shevlin R, Nisbet A. Correction of Geometric Image Distortion Using FPGAs. Optical Metrology, Imaging, and Machine Vision Conference, Galway, IRL, SPIE-Int'l Society for Optical Engineering, 2002
[46] Flavio, L.C. Padua, Guilherme A.S. Permeira, etc. Improving Processing Time of Large Images by Instruction Level Parallelism. Workshop on Parallel and Distributed Systems, Punta Arenas, Chile, 2001
[47] Catmull E., Smith A.R. 3-D transformations of images in scanline order. Computer Graphics, (SIGGRAPH'80 Proceedings), July 1980, 14(3): 279-285
[48] Wolberg G, Boult T.E. Separable image warping with spatial lookup tables. ACM Computer Graphics, 1989, 23(3): 369-378
[49] Wolberg G., Massalin H.. Fast convolution with packed lookup tables. Graphics Gems IV, Ed. By P. Heckbert, Academic Press, 1994: 447-464
[50] Fant K.M. A nonaliasing, real-Time spatial transform technique. IEEE Computer Graphics and Applications, 1986, 6(1): 71-80
[51] Meinds K., Barenbrug B. Resample hardware for 3D graphics. SIGGRAPH/Eurographics Graphics Hardware workshop, 2002: 17-26
[52] Warpenburg M. R., Siegel L.J. SIMD Image Resampling. IEEE Transactions on Computers, 1982, 31(10): 934-942,
[53] Jenq J.F., Sahni S. Reconfigurable mesh algorithms for image shrinking, expanding, clustering and template matching. In Proceedings of the Fifth International Parallel Processing Symposium, 1991: 648-656
[54] Wittenbrink CM., Somani A.K. 2D and 3D optimal image warping. Journal of Parallel and Distributed Computing. 1995, 25(2): 197-208
[55] Sylvain C.V., Serge M. A load-balanced algorithm for parallel digital image warping. International journal of pattern recognition and artificial intelligence, 1999, 13(4): 445-463
[56] Musick R., Catlett J., Russell S. Decision theoretic subsampling for induction on large database. In Utgoff, Proceedings of the Tenth International Conference on Machine Learning. San Francisco, CA. Morgan Kaufmann. 1992: 212-219
[57] Shafer J., Agrawal R., Mehta M. SPRINT: A scalable parallel classifier for data mining. In proceedings of the Twenty-second VLDB Conference. San Francisco, CA. Morgan Kaufmann. 1996:544-555
[5
[58] Ruggieri S. Efficient C4.5. IEEE Transactions on Knowledge and Data Engineering, 2001, 14(2): 438-444
[59] Chan RK., Stolfo S.J. Learning arbiter and combiner trees from partitioned data tbr scaling machine learning. In Proceedings of the First International Conference on Knowledge Discovery and Data Mining. Menlo Park, CA. AAAI Press. 1995:39-44
[60] Bowyer K., Chawla N., Moore T., Hall L., Kegelmeyer W. A parallel decision tree builder for mining very large visualization datasets. In Proceedings of IEEE-System, Man and Cybernetics, 2000
[61] Zaki M.J., Ho C.T, Agrawal R. Parallel Classification for Data Mining on Shared-Memory Multiprocessors. IBM Technical Report, 1998:198-205
[62] Joshi M.V., Karypis G, Kumar V. ScalParC: A New Scalable and Efficient Parallel Classification Algorithm for Mining Large Datasets. In Proceedings of the International Parallel Processing Symposium, 1998
[63] Caccetta RA., Campbell N.A., West G.A. A massively parallel implementation of an image classifier. In Proceedings of the 16th Asian Conference on Remote Sensing, 1995
[64] 刘勇卫，贺学鸿．遥感精解．测绘出版社，1993年12月第一版，170-220
[65] Wharton, S.W. A spectral-knowledge-based approach for urban land-cover discrimination, IEEE Transactions on Geoscience and Remote Sensing, May 1987, GE-25(3): 272-282
[66] Serpico S.B., Roli F. Classification of multisensor remote-sensing images by structured neural networks. IEEE Transactions on Geoscience and Remote Sensing, 1995, 33(3): 562-578
[67] Roli F. Multisensor image recognition by neural networks with understandable behavior. International journal of pattern recognition and artificial intelligence, 1996, 10(8): 887-917
[68] Setiono R. Extracting Rules from Neural Networks by Pruning and Hidden-Unit Splitting. Neural Computation, 1997, 9(1): 205-225
[69] Craven M.W. Extracting comprehensible models from trained neural network. Ph. D thesis, University of wisconsin, madison, 1996
[70] Bay, S.D. UCI KDD Archive [http://kdd.ics.uci.edu] , 1999
[71] Jiang Y.H., Zhao Q.L., Yang X.J. A general coding method for error-correcting output codes. LNCS 3056/PAKDD2004, Sydney Australia, May 26-28, 2004, 已录用
[72] 蒋艳凰，杨学军．基于搜索输出编码的简单贝叶斯分类方法．国防科学技术大学学报，已录用
[73] Quinlan, J.R. C4.5: Programs for Machine Learning. San Mateo, CA: Morgan Kaufmann Publishers, Inc., 1993
[7
[74] Breiman L., Friedman J.H., Olshen R.A., Stone C.J. Classification and Regression Trees. Wadsworth Ⅰ nternational Group, 1984
[75] Rumelhart D.E., Hinton G., Williams R. Learning Internal Representations through Error Propagation. In Rumelhart D.E., McClelland J.L. & the PDP research Group, (Eds.), Parallel Distributed Processing: Experiments in the Microstructure of Cognition, Vol. 1. Cambridge: MIT Press, 1986
[76] Freund Y., Schapire R.E. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Science. 1997, 55(1): 119-139
[77] Amari S., Wu S. Improving support vector machine classifiers by modifying kernel functions. Neural Networks, 1999, 12(6): 783-789
[78] Allwein, R. E. Schapire, Singer Y. Reducing multiclass to binary: A unifying approach for margin classifier. Journal of Machine Learning Research, 2000, 1:113-141
[79] Hastie T., Tibshirani R. Classification by pairwise coupling. The Annals of Statistics, 1998, 26(2): 451-471
[80] Sejnowski T.J., Rosenberg C.R. Parallel networks that learn to pronounce English text. Journal of complex systems, 1987, 1(1): 145-168
[81] 肖国镇，卿斯汉．编码理论．北京：国防工业出版社，1993
[82] Bose R.C, Ray-Chaudhuri D.K. On a class of error-correcting binary group codes. Information and Control, 1960, 3(1): 68-79
[83] Crammer K., Singer Y. On the learnability and design of output codes for multiclass problems. In Proceedings of the Thirteenth Annual 'Conference on Computational Learning Theory, 2000:35-46
[84] Dietterich T. Machine learning research: four current directions. AI Magazine, 1997, 18(4): 97-136
[85] Hansen L., Salamon P. Neural network ensembles. IEEE Trans. Pattern Analysis and Machine Intell. 1990, 12: 993-1001
[86] 蒋艳凰，杨学军．多层组合分类器研究．计算机工程与科学，已录用
[87] Kong E.B, Dietterich T.G. Error-Correcting Output Coding Corrects Bias and Variance. In proceedings of the 12th International Conference on Machine Learning. In S. Prieditis and S. Russell, eds. 1995:313-321
[88] Gama J. Combining classification algorithm. Ph.D. Thesis, 1999
[89] Kohavi R., Wolpert D.H. Bias plus variance decomposition for zero-one loss functions. In Proceedings of the 13th International Conference on Machine Learning. Morgan Kaufmann, 1996:275-283
[90] Langley P. Induction of recursive Bayesian classifiers. In Proceedings of the 6th European Conference on Machine Learning. Lecture Notes in Artificial Intelligence, 1993, 667:153-164
[9
[91] 周志华，陈世福．神经网络集成．计算机学报，2002，25(1)：1-8
[92] Ting, K.M., Zheng, Z.J. Improving the performance of boosting for naive Bayesian classification. In Proceedings of Pacific-Asia Conference on Knowledge Discovery and Data Mining, Berlin: Springer-Verlag. 1999:296-305
[93] 吴翊，李永乐，胡庆军．应用数理统计．国防科技大学出版社，1997
[94] Cestnik B. Estimating probabilities: a crucial task in machine learning. In proceedings of ECAI 90, Stockholm, 1990
[95] Dzeroski S., Cestnik B., Petrovski I. Using the m-estimate in rule induction. Journal of Computing and Information Technology, 1993, 1(1): 37-46
[96] Cybenko G. Continuos valued neural networks with two hidden layers are sufficient. Technical Report, Department of Computer Science, Tufts University, Medford, MA, 1988
[97] Jimenez D. Dynamically weighted ensemble neural networks for classification. In proceedings of IJCNN98, Vol. 1. Anchorage, AK, IEEE Computer Society Press, Los Alamitos, CA, 1998:753-756
[98] Krogh A., Vedelsby J. Neural network ensembles, cross validation, and active learning. In Tesauro G., Touretzky D., Leen T. (Eds.), Advances in Neural Information Processing Systems 7, Denver, CO, MIT Press, Cambridge, MA, 1995:231-238
[99] Lippman R.P. An introduction to computing with neural nets. IEEE ASSP Magazine, 1987, 2:4-22
[100] Kanellopoulos I., Wilkinson G.G. Strategies and best practice for neural network image classification. International Journal of Remote Sensing, 1997, 18(4): 711-725
[101] Hwang J.N., Lay S.R., Kiang R. Robust construction neural networks for classification of remotely sensed data. In Proceedings World Congress on Neural Network. Portland, July 11-15, 1993, 4:580-584
[102] 蒋艳凰，周海芳，杨学军．一种结构化神经网络及其特性分析．计算机研究与发展．2002，39(增刊)：291-297
[103] 蒋艳凰，周海芳，杨学军．基于纠错编码的CSNN及其在遥感图像分类中的应用．计算机研究与发展，2003，40(7)：918-924
[104] Hertz J., Krogh A., Palmer R.G. Introduction to the Theory of Neural Computation. Reading, MA: Addison Wesley. 1991
[105] Peng C.H., Wen X.Z. Recent Applications of Artificial Neural Networks in Forest Resource Management: An Overview. Environmental Decision Support Systems and Artificial Intelligence. Menlo Park, CA: AAAI press, 1999:15-22
[106] 袁曾任．人工神经元网络及其应用．北京：清华大学出版社．1999
[107] Shavlik J.W., Mooney R.J., Towell G.G. Symbolic and neural learning algorithms: an experimental comparison. Machine Learning. 1991, 6(1): 111-143
[108] Chandrasekaran B., Goel A. From numbers to symbols to knowledge structures: Artificial intelligence perspective on the classification task. IEEE Trans, Syst, Man Cybernet. 1988, 18(3): 415-424
[109] Drago G.P, Ridella S. Statistically controlled activation weight initialization (SCAWI). IEEE Trans Neural Networks. July 1992, 3:627-631
[110] Darken C, Moody J. Note on learning rate schedules for stochastic optimization. Neural Information Processing Systems. Lippmann R.P, Moody J. E., Touretzky, D. S. (Editors) 1991:832-838
[111] Salomon R. Improved convergence rate of back-propagation with dynamic adaptation of the learning rate. Lecture Notes in Computer Science, RPSN Ⅰ. Dortmund, 1990:269-273
[112] Tollenaere T. SuperSAB: fast adaptive backpropagation with good scaling properties. Neural Networks. 1990, 3:561-573
[113] Jacobs R.A. Increased rates of convergence through learning rate adaptation. Neural Networks, 1988, 1(4): 295-308
[114] Riedmiller M., Heinrich B. A direct adaptive method for faster backpropagation learning: the RPROP algorithm. In Proceedings of the IEEE Intemational Conference on Neural Networks (ICNN). San Francisco, 1993:586-591
[115] Riedmiller M. Rprop-description and implementation details. Technical Report. Institut fur Logic Konplexitat and Deduktionssyteme University of Karlsruhe FRG, January 1994
[116] Igel C., Husken M. Improving the Rprop learning algodthrn. In Proceedings of the Second International Symposium on Neural Computation, NC'2000, ICSC Academic Press, 2000:115-121
[117] Simon H.A. The Science of the Artificial. Cambridge, MA: MIT. Press, 2nd edition, 1981
[118] 张朝晖，陆玉昌，张钹．发掘多值属性的关联规则．软件学报，1998，9(11)：801-805
[119] 张朝晖．数据采掘的多策略机器学习方法．[博士论文] ．清华大学计算机系，北京，1998
[120] Daugherty J., Kohavi R., Sahami M. Supervised and unsupervised discretization of continuous feature. In Proceedings of Twelfth International Conference on Machine Learning, San Francisco: Morgan Kaufrnann, 1995:194-202
[121] Jiang Y.H., Zhou H.E, Yang X.J. Constructing Decision Tree with Continuous Attributes for Binary Classification. IEEE Conference ICMLC2002, Beijng, China, November 2002, 4(2): 617-622
[122] Jiang Y.H., Yang X.J., Zhao Q.L. Constructing Decision Tree with High Intelligibility. Journal of Software. 2003, 14(12): 1996-2005
[123] Liu H., Hussain R, Tan C.L., Dash M. Discretization: an enabling technique. Data Mining and Knowledge Discovery, 2002, 6(4): 393-423
[124] Zemke S., Rams M. Multivariate Feature Coupling and Discretization. FEA-2003, US 2003
[125] Bay S. D. Multivariate discretization for set mining. Knowledge and Information Systems, 2001, 4(3): 491-512
[126] Holte R.C. Very simple classification rules perform well on most commonly used datasets. Machine Learning, 1993, 11: 63-90
[127] Quinlan J.R. Induction of decision trees. Machine Learning, 1986, 1: 81-106
[128] Quinlan J.R. Improved use of continuous attributes in C4.5. Journal of Intelligence Research, 1996,4:77-90
[129] Catlett J. On changing continuous attributes into ordered discrete attributes. Proceeding of the Fifth European Working Session on Learning, Heidelberg: Springer-Verlag, 1991: 164-178
[130] Fayyad U.M., Irani K.B. Multi-interval discretization of continuous-valued attributes for classification learning. Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence. San Francisco, CA: Morgan Kaufman, 1993: 1022-1027
[131] Fayyad U.M., Irani K.B. Discretizing continuous attributes while learning bayesian networks. In Proc. Thirteenth International Conference on Machine Learning. Morgan Kaufmann, 1996: 157-165
[132] Cerquides J., Mantaras R.L. Proposal and empirical comparison of a parallelizable distance-based discretization method. In KDD97: 3rd International Conference on Knowledge Discovery and Data Mining, 1997: 139-142
[133] Mantaras R.L. A distance based attribute selection measure for decision tree induction. Machine Learning, 1991: 103-115
[134] Ho K.M., Scott P.D. Zeta: A global method for discretization of continuous variables. In KDD97: 3rd International Conference on Knowledge Discovery and Data Mining, 1997: 191-194
[135] Kerber R. ChiMerge: Discretization of numeric attributes. In Proc AAAI-92, Ninth National Conference on Artificial Intelligence, pages 123-128. AAAI Press/The MIT Press, 1992
[136] Liu H., Setiono R. Feature selection and discretization. IEEE Transactions on Knowledge and Data Engineering, 1997, 9:1-4
[137] Wallace C.S., Patrick J.D. Coding decision trees. Machine Learning, 1993, 11:7-22
[138] Auer P., Holte R.C, Maass W. Theory and application of agnostic PAC-learning with small decision trees. In: Prieditis, A., Russell, S. Eds. Proceedings of the 12th International Conference on Machine Learning. San Francisco: Morgan Kaufmann Publishers, Inc., 1995: 21-29
[1
[139] Elomaa T., Rousu J. General and Efficient Multisplitting of Numerical Attributes. Machine Learning, 1999, 36:1-49
[140] Rousu J. Efficient range partitioning in classification learning. PhD thesis, Series of publication A, Report A-2001-1. University of Helsinki, 2001
[141] 史忠植．知识发现．北京：清华大学出版社．2002
[142] Quinlan J.R. Simplifying decision tree. Intemational Journal of Man-Machine Studies, 1987, 27:221-234
[143] Kenneth R．Castleman著．数字图像处理(Digital image processing)．北京：清华大学出版社．2003
[144] Buyya R. High Performance Cluster Computing: Architectures and Systems, volume 1. Prentice Hall PTR, Upper Saddle River, NJ, 1999
[145] 林凡．集群的可扩展性及其分布式体系结构(上和下)．IBM developerWorks中国网站．
http://www-900.ibm.com/developerWorks/cn/linux/cluster/clustersystem/base/part1/index.shtml
http://www-900.ibm.com/developerWorks/cn/linux/cluster/cluster_System/base/part2/indexl.shtml
http://www-900.ibm.com/developerWorks/cn/linux/cluster/cluster_system/base/part2/index2.shtml
[146] Jiang Y.H., Yang X.J., Dai H.D., Yi H.Z. A Distributed Parallel Resampling Algorithm for Large Images. LNCS 2834/APPT2003, Xiamen, China, September 17-19, 2003:608-618
[147] 蒋艳凰，杨学军，易会战．基于Cluster环境的大图像并行重采样技术研究．计算机研究与发展，已录用
[148] 蒋艳凰，杨学军，易会战．卫星遥感图像并行几何校正算法研究．计算机学报，已录用
[149] Paul S. Heckbert. Survey of texture mapping. IEEE Computer Graphics and Application, November 1986:207-212
[150] Beier T., Neely S. Feature-based image metamorphosis. Computer Graphics, 1992, 26(2): 35-42.
[151] Chen S.E. Quicktime VR - an image-based approach to virtual environment navigation. Computer Graphics (SIGGRAPH 95), August 1995: 29-38
[152] Thévenaz P., Blu T., Unser M. Image interpolation and resampling. Handbook of Medical Imaging, Processing and Analysis, I.N. Bankman, Ed., Academic Press, San Diego CA, USA, 2000, 393-420
[153] Breene L.A., Bryant J. Image warping by scanline operations. Computers & Graphics, 1993, 17(2): 127-130
[154] Heckbert RS. Fundamentals of texture mapping and image warping. Technical Report, CS Division, EECS Department, UC Berkeley, May 1989
[155] Li Z-C. Analysis of discrete techniques for image transformation. Numerical Algorithms, 1996, 13(3-4): 225-263
[156] Zhang Y. A fuzzy approach to digital image warping. IEEE Computer Graphics & Applications, July 1996, 16(4): 34-41
[157] Foley J.D., Dam A.V., Feiner S.K., Hughes J.F. Computer Graphics, Principles and Practice, 2nd edition, Addison-Wesley, Reading, Massachusetts, 1990
[158] Keys R.G. Cubic convolution interpolation for digital image processing. IEEE transactions on Acoustics, Speech, and Signal Processing, 1981, 29(6): 1153-1160
[159] Lee C., Hamdi M. Parallel image processing applications on a network of workstations. Parallel Computing, 1995, 21:137-160
[160] 周海芳．遥感图像并行处理算法的研究与应用(博士论文)．长沙：国防科学技术大学，2003．10
[161] Hillis W.D., Steele G.L., Jr. Data parallel algorithms. Communcations of ACM (Special issue on parallism), 1986, 29(12): 1170-1183
[162] Serge M. and Yves R. Elastic load balancing for image processing algorithms. In HP Zima, editor, Parallel Computation, LNCS 591, Springer Verlag, 1992:438-451
[163] PCI Geomatics主页. http://www.pcigeomatics.com/
[164] ERDAS公司主页. http://www.erdas.com/ERDASredirect/index2.html
[165] RSI公司主页. http://www.rsinc.com/envi/index.cfm
[166] RSIM主页. http://159.226.117.45/rsim/index.htm
[167] Thain D., Livny M. The ethernet approach to grid computing. In the Proceedings of the Twelfth IEEE Symposium on High Performance Distributed Computing, Seattle, WA, 2003
[168] Nicholas M. Short, Sr. Remote sensing and image interpretation & analysis, http://mercator.upc.es/tutorial/table.html.