|本期目录/Table of Contents|

[1]叶 珍,白 璘.局部保护降维与高斯混合模型的高光谱图像分类[J].工业仪表与自动化装置,2017,(04):3-6.
　YE Zhen,BAI Lin.Hyperspectral image classification based on locality-preserving dimension reduction and Gaussian mixture model[J].Industrial Instrumentation & Automation,2017,(04):3-6.
点击复制

局部保护降维与高斯混合模型的高光谱图像分类

《工业仪表与自动化装置》[ISSN:1000-0682/CN:61-1121/TH]

卷:
期数:
2017年04期
页码:
3-6
栏目:
出版日期:
2017-08-15

文章信息/Info

Title:
Hyperspectral image classification based on locality-preserving dimension reduction and Gaussian mixture model
文章编号:
1000-0682（2017）04-0000-00
作者:
叶 珍; 白 璘
（长安大学 电子与控制工程学院，西安 710064）
Author(s):
YE Zhen;  BAI Lin
(School of Electronics and Control Engineering, Chang’an University, Xi’an 710064, China)
关键词:
高斯混合模型; 局部保护投影; 局部保护非负矩阵分离; 高光谱图像分类
Keywords:
Gaussian mixture model;  locality preserving projection;  locality preserving non-negative matrix factorization;  hyperspectral image classification
分类号:
TP751
DOI:
-
文献标志码:
A
摘要:
高光谱图像具有高谱间分辨率和低空间分辨率的特点，传统的分类方法难以得到较高的分类精度。针对该问题，该文研究了两种局部保护降维法-局部保护投影(LPP)和局部保护非负矩阵分离(LPNMF)对高光谱图像降维，这两种方法能很好地保护输入空间相邻像素间的局部特征；由于高光谱图像各类间的统计分布多为复杂的多模型结构，文中采用高斯混合模型(GMM)分类器对降维后的数据进行分类。实验结果表明，将局部保护降维与高斯混合模型相结合的高光谱图像分类算法不但在小样本情况下能有效地提高分类精度，而且在背景像素混合的情况下和高斯白噪声环境中具有一定的鲁棒性。
Abstract:
Since hyperspectral imagery has the characteristic of low spatial resolution and hundreds of spectral bands for each pixel, the classification accuracy of traditional classification methods is unsatisfactory.For this problem, two locality-preserving dimension reduction methods-locality preserving projection(LPP) and locality preserving nonnegative matrix factorization(LPNMF) are studied in this paper for preserving the local structure of neighboring samples.The statistical distribution of hyperspectral image classes is often a complicated multimodal structure. Therefore，classifiers based on Gaussian mixture model(GMM) are hence a natural fit for hyperspectral data. Experimental results show that the algorithms combined locality-preserving dimension reduction methods with Gaussian mixture model can be effective in small training-samples-size situation, as well as provides outstanding robustness under background pixel-mixing conditions and strong Gaussian noise environments.

参考文献/References:

[1] Ye Z, Prasad S, Li W, et al. Classification Based on 3D DWT and Decision Fusion for Hyperspectral Image Analysis[J]. IEEE Geoscience and Remote Sensing Letters, 2014, 11(1): 173-177.

[2] Prasad S, Bruce L M. Limitation of principal component analysis for hyperspectral target recognition[J]. IEEE Geoscience and Remote Sensing Letters, 2008, 5(4): 625-629.
[3] Duda R O, Hart P E, Stork D G. Pattern Classification[M]. 2nd ed. New York: John Wiley & Sons, Inc., 2001: 215-219.
[4] Li W,Prasad S,Ye Z,et al. Locality-preserving discriminant analysis for hyperspectral image classification using local spatral information[C]//Geoscience and Remote Sensing Symposium. Munich, Germany, 2012: 4134-4137.?
[5] Cai D,He X,Wang X,et al.Locality preserving nonnegative matrix factorization[C]// Pasadena, California: The International Joint Conference on Artificial Intelligence, 2009: 1010-1015.
[6] Berge A and Solberg A H S. Structured Gaussian components for hyperspectral imgae classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(11): 3386-3396.
[7] Jia S,Qian Y. Constrained nonnegative matrix factorization for hyperspectral unmixing[J].IEEE Transactions on Geoscience and Remote Sensing, 2009, 47(1): 161-173.
[8] Lee D D, Seung H S. Learning the parts of objects by non- negative matrix factorization[J].Nature,1999,401(6755): 788-791.
[9] Li W, Prasad S, Fowler J E, et al. Locality-preserving nonnegative matrix factorization for hyperspectral image classification[C]// Munich,Germany: Geoscience and Remote Sensing Symposium, 2012: 4134-4137.
?[10] Vlassis N, Likas A. A greedy EM algorithm for Gaussian mixture learning[J].Neural Processing Letters, 2002, 15(1): 77-87.
[11] Dobigeon N,Moussaoui S, Coulon M, et al. Joint Bayesian endmember extraction and linear unmixing for hyperspectral imagery[J]. IEEE Transactions on Signal Processing, 2009, 57(11): 4355-4368.
[12] Pu H Y, Chen Z, Wang B, et al. A novel spatial-spectral similarity measure for dimensionality reduction and classification of hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2014, 52(11): 7008-7022.

相似文献/References:

[1]吴文昭.基于GMM聚类的鲁棒性i向量说话人确认[J].工业仪表与自动化装置,2017,(04):55.
　WU Wenzhao.Speaker verification robust speaker recognition based on i-vector and GMM clustering[J].Industrial Instrumentation & Automation,2017,(04):55.

备注/Memo

备注/Memo:
收稿日期：2016-10-13
基金项目：国家自然科学基金项目（41601344,61601059）；中央高校基本科研业务费专项资金项目（310832163402, 310832161001）
更新日期/Last Update: 1900-01-01