摘要
遥感图像分类一直是遥感研究领域的重要内容,如何解决多类别图像的分类识别并满足一定的精度是遥感图像研究中的一个关键问题,具有十分重要的意义。传统的遥感图像自动分类识别主要采用决策理论(或统计)方法进行分类,但是,由于遥感图像本身的空间分辨率低,以及“同物异谱”、“异物同谱”现象的存在,往往出现较多的错分、漏分情况,导致分类精度不高。人工神经网络具有高度并行处理能力、自适应能力、非线性映射能力、泛化能力,使得人工神经网络在遥感图像分类应用研究中提供了新的方法。
传统的遥感图像分类法主要分为监督分类法和非监督分类法,本文首先简要概括了几种常用的监督分类和非监督分类法。介绍了人工神经网络的结构原理、学习规则和在遥感图像分类中典型的神经网络模型,详细的分析了BP神经网络模型的遥感图像分类,并优化了基本的BP算法。重点讨论了自组织神经网络在遥感图像分类中的应用,针对自组织神经网络算法的学习率收敛慢做了改进,采用幂函数收敛。本文以Matlab为平台,构建BP神经网络和改进的自组织神经网络,
对北京市海淀区遥感影像进行实验,并对实验结果进行了分析、比较。结果表明,改进的自组织神经网络分类方法分类精度远高于传统的遥感图像分类法和BP神经网络分类法分类精度,采用改进自组织神经网络模型的遥感图像分类法是有效的。
Remote sensing(RS)image classification is always a pivotal part of remote sensing study. How to improve the accuracy of RS interpretation is a urgent problem in RS application. The automatic classification of traditional RS image is principal used in decision-making theory (or statistical) methods to classify. However, the existence of the phenomenon about RS images of low spatial resolution is appeared as well as "synonyms spectrum", "foreign body spectrum with", more often at fault, leakage points, which induce the low classification accuracy .The artificial neural network (ANN) has high degree of parallel processing, adaptive capacity, non-linear mapping capabilities, generalization ability. So artificial neural networks in remote RS in classifications applied research provides a new approach.
Traditional RS classification is divided into supervised classification and unsupervised classification. At first, the paper briefly introduced several commonly used classification of supervised and unsupervised methods. Then we introduce the ANN structure and learning rules , as well as typical neural network models in remote sensing image classifications. The BP neural network model of remote sensing image classification is analysised detailed, and the basic algorithm for BP is optimized.Lastly,self-organizing feature neural network(SOFM) in remote sensing image classification is emphasized .In this paper,the power function convergence is used to solve the slow rate of SOFM learning algorithm convergence.
Based on MATLAB as a platform, we build BP neural network and improved SOFM neural network, making Haidian District, Beijing remote sensing imaging as experiments object. We analysis and comparison the experiments results. The experiment shows that the improvement of SOFM neural network classification accuracy is higher than the traditional classification of RS image and BP neural network classification accuracy. So the improved SOFM neural network model of RS image classification is valid.
引文
1 Chi Hau Chen, Pei-Gee Peter Ho. Statistical Pattern Recognition in Remote Sensing .Pattern Recognition, 2008,41(9): 2731-2741
2 P. Townsend, A. Walsh. Remote Sensing off Orested Wetlands:Aplication of Multitemporal and Multi Speetra lsatellite Imagery to Determine Plant Community Composition and Structure in Southeastern USA.Plant Ecology,2001,(157):129-149
3 O.Yashon,S. Josaphat, Ryutaro Tateishi.Multiscale Remote Sensing Data Segmentation and Post-segmentation Change Detection Based on Logical Modeling: Theoretical Exposition and Experimental Results for Forestland Cover Change Analysis. Computers & Geosciences,2008,34(7): 715-737
4 Yushun Zhai, E. Julian, Boggess III, et al. Soil Texture Classification with Artificial Neural Networks Operating on Remote Sensing Data. Computers and Electronics in Agriculture,2006,54(2): 53-68
5马静.遥感图像分类方法的研究. [山东科技大学工学硕士论文].2003:10-15
6 C. Corbane, F. Jacob, J. Albergel, et al. Remote Sensing of Soil Surface Characteristics from a Multiscale Classification Spproach. CATENA,In Press, Corrected Proof, 2008,6:44-47
7 Wang Runsheng. Image Understanding . Changsha: The National Defence Science and Technology University Press, 1995:80-83
8 P. Gong. Integrated Analysis of Spatial Data from Multiple Sources: Using Evidential Reasoning and Artificial Neural Network Techniques for Geological Mapping. Phot- ogrammetric Engineering & Remote sensing,1996,62(5): 513-523
9 Oliver Sonnentag, J. Canty. Image Analysis. Classification and Change Detection in Remote Sensing (With Algorithms for ENVI/IDL),CRC Press, Boca Raton, Hardbound. Computers & Geosciences,2008,34(1):93-94
10 Zhengjun Liu, Aixia Liu, Changyao Wang, et al. Evolving Neural Network Using RealCoded Genetic Algorithm (GA) for Multispectral Image Classification. Future Generation Computer Systems,2004,20(7):1119-1129
11马少平.人工神经网络在遥感图像分类中的应用研究与开发.[中国地质大学工学硕士论文].2006:13-18
12 Awa Niang, Lidwine Gross, Sylvie Thiria, et al. Automatic Neural Classification of Ocean Colour Reflectance Spectra at the Top of the Atmosphere with Introduction of Expert Knowledge.Remote Sensing of Environment,2003,86(2): 257-271
13 Liangpei Zhang, Ke Wu, Yanfei Zhong, et al. A New Sub-pixel Mapping Algorithm Based on a BP Neural Network with an Observation Model.Neuro Computing, 2008,71(10): 2046-2054
14赵春霞,钱乐祥.遥感影像监督分类与非监督分类的比较,河南大学学报(自然科学版),2004,34(3):90-93
15骆剑承等.遥感地学智能图解模型支持下的土地覆盖土地利用分类.自然资源学报, 2001,16(2):179-183
16 M. Lillesand, Ralph, W. Kiefer.遥感与图像解译(彭望碌,余先川,译者),第4版.北京:电子工业出版社,2003:113-134
17马建文,李启青,哈斯巴干.遥感数据智能处理方法与程序设计.北京:科学出版社, 2005:48-63
18王任华,霍宏涛,游先祥.人工神经网络在遥感图像森林植被分类中的应用.北京林业大学学报,2003,7(4):1-5
19 A. Niang, F. Badran, C. Moulin, et al. Retrieval of Aerosol Type and Optical Thickness over the Mediterranean from Sea WiFS Images Using an Automatic Neural Classification Method.Remote Sensing of Environment,2006,100(1): 82-94
20潘东晓,虞勤国,赵元洪.遥感图像的神经网络分类法.国土资源遥感,1996(,3):49-55
21 Fu Limin. Neural Networks in Computer Intelligence. McGraw Hill, Inc. 1994
22陆荣建.分形分维和神经网络在遥感影像分析中的应用研究.[北京大学工学硕士论文].2000:17-21
23 J. Benediktsson, P. Swain, O. Ersoy. Neural Network Approaches Versus Statistical Methods in Classification of Multisource Remote Sensing Data.IEEE Transactions on 59Geoscience and Remote Sensing,1990,28(4): 540-552.
24 Wilkinson GA. Review of Current Issues in the Integration of GIS and Remote Sensing Data .Int. J. Geographical Information Systems,1996,10(1): 85-101
25 H. Mural, S. OMATU. Remote Sensing Image Analysis Using a Neural Network and Knowledge Based Processing .INT. J. Remote Sensing,1997,18(4): 811-828
26曾鹏,徐建军,吴玲达.卫星数字图像的智能处理研究.系统仿真学报,2005,17(7):32-35
27曹广真.基于Landsat ETM+和ERS-2 SAR图像融合对城区地特征分类的BP-ANN/GA混合算法.电子与信息学报,2007.7(28):1153-1159
28哈斯巴干.神经网络及其组合算法的遥感数据分类研究.[中国科学院研究生院工学博士论文].2003:45-56
29钱晓东.基于神经网络等技术的数据与文本聚分类研究.[天津大学工学博士论文]. 2005:25-28
30 Jixian Zhang, Yonghong Zhang. Remote sensing research issues of the National Land Use Change Program of China.ISPRS Journal of Photogrammetry and Remote Sensing,2007,62(6):150-156
31吐热尼古丽,张晓帆.基于人工神经网络的遥感图像分类研究.长春师范学院学报(自然科学版),2006,(2):81-84
32 L. Castellana, A. D’Addabbo, G. Pasquariello. A Composed Supervised/Unsupervised Approach to Improve Change Detection from Remote Sensing. Pattern Recognition Letters,2007,28(4): 405-413
33 R. Van Kane, R. Alan Gillespie, Robert McGaughey, et al. Franklin.Interpretation and Topographic Compensation of Conifer Canopy Self-shadowing.Remote Sensing of Environment,2008,112(10):3820-3832
34 X.-H.Liu. An Experimental Study on Spectral Discrimination Capability of a Back Propagation Neural Network Classifier.INT.J.Remote Sensing,2003,24(4):673-685
35董延华.超光谱遥感图像处理关键技术研究.[哈尔滨理工大学工学硕士论文]. 2006:25-33
36 Wang Xiaoping, Ding Xiang, Dou Aixia. Study on Model Base System of EarthquakeDamage Extraction from RS in ENVI/IDL Enviroment.IEEE International Geoscience and Remote Sensing,2004,(4):2252-2255
37 A. Benediktsson, H. Swain, K. Ersoy. Neural Network Approaches Versus Statistical Methods in Classification of Multi Source Remote Sensing Data.IEEE Transactions on Geosciences and Remote Sensing,1990,28(4):540-552
38 J. Zhou, L. Daniel. Civco.Using Genetic Learning Neural Networks for Spatial Decision Making in GIS.Photogram Metric Engineering & Remote Sensing. 1996,62(1): 1287-1295
39贾永红.人工神经网络在多源遥感影像分类中的应用.测绘通报,2000,(7):7-8
40丁建丽,塔西甫拉提·特依拜,刘传胜.人工神经网络模型及其在遥感中的应用.新疆大学学报(理工版),2001,18(3):269-276
41段晓芹,对LANDSAT7 ETM+遥感图像分类识别技术的研究与应用.[内蒙古大学工学硕士论文].2006:35-43
42张宝光.人工神经网络在遥感数字图像分类处理中的应用.国土资源遥感, 1998,(1):21-27
43 P. Atkinson, J. Cutler, H. Lewis. Mapping sub Pixel Proportional Land Cover with AVHRR Imagery .INT. J. Remote Sensing,1997,18(4): 917-935
44刘旭升等.基于BP神经网络的森林植被遥感分类研究.林业资源管理, 2005,(1):52-54
45罗扬帆.基于BP神经网络的遥感影像分类研究.[北京林业大学工学硕士论文]. 2007:56-63
46戴永伟,雷志勇.BP网络学习算法研究及其图像模式识别应用.计算机与现代化, 2006,(11):67-70
47 Kohonen T, et al. Engineering Application of the Self Organizing Map.PIEEE, 1996,84(10):1358-1384
48毛建旭.王耀南.基于神经网络的遥感图像分类.测控技术,2001,20(5):29-33
49郭志强,蔡嵩.彩色遥感图像分类算法及Matlab实现.武汉理工大学学报,2006, 28(1):108-111
50林崇献,马隆文.BP网络及其在遥感图像分类中的应用.南方国土资源,2005,(4):25-27
51许将军.高光谱遥感图像特征提取及分类研究—基于K-L变换据支撑矢量技术.[电子科技大学工学硕士论文].2004:56-59
52李祚泳.用B-P神经网络实现多波段遥感图像的监督分类.红外与毫米波学报, 1998,17(2):153-156
53王银辉,熊忠阳.使用BP网络改进K-means聚类效果.计算机科学,2006,(3):194-196
54骆剑承,周成虎,杨艳.人工神经网络遥感影像分类模型及其与知识集成方法研究.遥感学报,2001.3(5):122-128
55修丽娜,刘湘南.人工神经网络遥感分类方法研究现状及发展趋势探析.遥感技术与应用,2003.10(18):339-345
56隋青美,王正欧.改进的自组织神经网络算法的收敛性(II).电机与控制学报, 2002.12(4):236-238
57王升明,李淼.一种基于改进的自组织特征映射网络的文档聚类方法.计算机工程与应用,2005,(3):167-169
58杜华强,范文义.Matlab自组织神经网络在遥感图像分类中的应用.东北林业大学学报,2003,(7):51-53
59熊祯,郑兰芬,童庆禧.分层神经网络分类算法.测绘学报,2000,29(3):229-234
60潘峰,刘文予,朱光喜.Matlab在图像处理与应用中的研究.计算机应用研究, 1999,(12):73-75
61孙家柄,舒宁,关泽群编著.遥感原理、方法和应用.北京:测绘出版社,1997:146-161
62高玉龙,张中兆.基于改进的自组织映射神经网络的调制方式识别分类器.四川大学学报(工程科学版),2006,9(38):143-147
63王山东,赵书河.基于改进的自组织映射网络的遥感影像融合研究.计算机应用与软件,2004,10(21):80-82
64刘仁钊,廖文峰.遥感图像分类应用研究综述.地理空间信息,2005,10(3):11-13
65吴轲,方强,张俊玲,等.基于改进Kohonen神经网络的遥感影像分类.测绘信息与工程,2007.2(32):47-49
66黎洪松.一种改进的自组织特征映射算法.中国民航学院学报,2006,24(1):36-38
67吕新彪.人工神经网络在地球科学中的应用总数.地址科技情报,1996,15(3):93-98