基于约束线性判别分析的非监督高光谱影像分类方法*

作者：张凯1，赵辽英1，厉小润2 日期：2009-08-18/span> 浏览：3969 查看PDF文档

基于约束线性判别分析的非监督高光谱影像分类方法*

张凯1，赵辽英1，厉小润2
(1.杭州电子科技大学 计算机应用技术研究所，浙江 杭州 310018； 2.浙江大学 电气工程学院，浙江 杭州 310027)

摘要：针对高光谱影像非监督分类问题，从特征提取的角度提出了一种用于高光谱混合像元分类的非监督约束线性判别分析算法（UCLDA）。该算法首先利用顶点成分分析(VCA)提取端元，然后用光谱角匹配方法(SAM)构造训练样本并基于约束线性判别分析(CLDA)进行特征提取，最后用最小距离法分类。整个算法实现了非监督分类。对模拟的高光谱数据和真实的遥感影像进行了仿真研究，研究结果表明，UCLDA略优于最小二乘光谱混合分析技术，但明显好于经典的光谱角匹配分类。
关键词：混合像元；端元提取；线性判别分析；非监督分类
中图分类号：TP751.1文献标识码：A文章编号：1001-4551(2009)08-0041-04

Unsupervised hyperspectral image classification method based on constrained linear discriminant analysis
ZHANG Kai1, ZHAO Liao-ying1, LI Xiao-run2
(1. Institute of Computer Application Technology, Hangzhou Dianzi University, Hangzhou 310018, China;
2. College of Electrical Engineering, Zhejiang University, Hangzhou 310027, China)
Abstract: Aiming at the problem of unsuperviesed hyperspectral image classification, a new unsupervised constrained linear discriminant analysis (UCLDA) approach to hyperspectral mixed pixel classification was introduced from the angle of feature extraction. Vertex component analysis (VCA) was used to extract the endmembers, then spectral angle mapping (SAM) was applied to get the training samples and let constrained linear discriminant analysis (CLDA) extract features of the hyperspectral image, and finally classification with least distance method was implemented. The new algorithm was unsupervised. The proposed algorithm was studied using simulated and real hyperspectral data. Experimental results show that the UCLDA is slightly better than least square spectral mixture analysis method and significantly superior to spectral angle mapping classification.
Key words: mixed pixels; endmember extraction; linear discriminant analysis; unsupervised classification
参考文献(References):
［1］赵英时.遥感应用分析原理与方法［M］.北京：科学出版社，2003.
［2］童庆禧,郑兰芬,王晋年，等.湿地植被成像光谱遥感研究［J］.遥感学报,1997,1(1):50-57.
［3］BAUGH W M, KRUSE F A, ATKINSON W W. Quantitative geochemical mapping of ammonium minerals in the southern cedar mountains, nevada, using airborne visible/ infrared imaging spectrometer (AVIRIS)［J］. Remote Sensing of Environment,1998,65(3):292-308.
［4］徐宏根,马洪超,李德仁.结合SOM神经网络和混合像元分解的高光谱影像分类方法研究［J］.遥感学报，2007,11(6):778-786.
［5］梁娜,何明一.基于遗传算法的混合像元快速分解及分类算法［J］.遥感技术与应用,2007,22(4):560-565.
［6］CHANG C I, REN H. Linearly Constrained Minimum Variance Beamforming for Target Detection and Classification in Hyperspectral Imagery［C］//In: Proceedings of IEEE International Geoscience and Remote Sensing Symposium., Hamburg: ［s.n.］,1999:1241-1243.
［7］DU Q, CHANG C I. A linear constrained distancebased discriminant analysis for hyperspectral image classification［J］. Pattern Recognition,2001,34(2):361-373.
［8］厉小润,赵光宙,赵辽英.约束线性描述分析与人脸识别［J］.系统仿真学报，2008，18(20):4937-4940.
［9］NASCIMENTO J M P, DIAS J M B. Vertex component analysis: a fast algorithm to unmix hyperspectral data［J］. IEEE Transactions on Geoscience and Remote Sensing,2005,43(4):898-910.
［10］童庆禧,张兵,郑兰芬.高光谱遥感：原理、技术与应用［M］.北京：高等教育出版社,2006.
［11］［作者不详］.ISGS Spectroscopy Lab［EB/OL］.［2007-09-25］.http://speclab.cr.usgs.gov.
［12］ABRAMS M, ASHLEY R, ROWAN L, et al. Mapping of hydrothermal alteration in the cuprite mining district, nevada, using aircraft scanner images for the spectral region 0.46 to 2.36 mm［J］. The Journal of Geology,1977,5(12):713-718.
［13］SWAYZE G. The hydrothermal and structural History of the Cuprite Mining District, southwestern Nevada: An integrated geological and geophysical Approach［D］. Boulder: University of Colorado,1997.
［14］CHANG C I, DU Q. Estimation of number of spectrally distinct signal sources in hyperspectral imagery［J］. IEEE Transactions on Geoscience and Remote Sensing,2004,42(3):608-619.