Hyperspectral Remote Sensing Image Classification Based on SSMFA and kNNS

WANG Li-zhi; HUANG Hong; FENG Hai-liang

doi:10.3969/j.issn.0372-2112.2012.04.026

您当前的位置：

首页 >

文章列表页 >

Hyperspectral Remote Sensing Image Classification Based on SSMFA and kNNS

更新时间：2025-07-16

- Hyperspectral Remote Sensing Image Classification Based on SSMFA and kNNS
- Acta Electronica Sinica Vol. 40, Issue 4, Pages: 780-787(2012)
- 作者机构：
  
  重庆大学光电技术及系统教育部重点实验室,重庆,400044
- 作者简介：
- 基金信息：
- DOI：10.3969/j.issn.0372-2112.2012.04.026
  CLC： TP751.1TP391.4
- Published：2012
- 稿件说明：
移动端阅览
WANG Li-zhi, HUANG Hong, FENG Hai-liang. Hyperspectral Remote Sensing Image Classification Based on SSMFA and kNNS[J]. Acta Electronica Sinica, 2012, 40(4): 780-787.
DOI：

WANG Li-zhi, HUANG Hong, FENG Hai-liang. Hyperspectral Remote Sensing Image Classification Based on SSMFA and kNNS[J]. Acta Electronica Sinica, 2012, 40(4): 780-787. DOI： 10.3969/j.issn.0372-2112.2012.04.026.

摘要

为了研究高光谱影像数据的维数约简和分类问题

提出了一种基于半监督边际费希尔分析(SSMFA)和

NNS的高光谱遥感影像数据分类算法.该方法利用有标记数据和无标记数据的信息获得数据的内在流形结构

通过SSMFA将高光谱数据从高维观测空间投影到低维流形空间

然后利用邻域内多个近邻点的信息通过

NNS分类器对低维空间中的数据进行分类.在Urban、Washington和Indian Pine数据集上的分类识别实验表明

该方法能够较为有效地发现高维空间中数据的内蕴结构

在每类随机选取4

8个有类别标记的样本10个无类别标记的样本的情况下

该方法的总体分类精度能够比MFA+

NNS提高0.8%~2.5%

比MFA+

NN提高2.8%~4.5%

比其他算法提高4.0%~7.0%

分类精度有了明显的提高.

Abstract

In order to explore dimensionality reduction and classification in hyperspectral remote sensing image

an algorithm based on semi-supervised marginal Fisher analysis(SSMFA) and k-nearest-neighbor simplex(

NNS) is proposed in this paper.First

the data are projected from a high-dimensional space onto low-dimensional space by SSMFA combined with the information of different classes.Then

classification is performed under the

NNS classifier by using a few neighbors from each class.The experimental results on the Urban data set

Washington DC Mall data set and Indian Pine data set show the effectiven

ess of the proposed algorithm

when i(i=4

8) labeled samples and 10 unlabeled samples of each class are randomly selected for training and 100 samples of each class for testing

the overall accuracy of our proposed algorithm is improved by 0.8%-2.5%

2.8%-4.5% and 4.0%-7.0%

respectively

as compared with MFA+

NNS

MFA+

NN and other methods.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A New Method for Steganalysis Techniques:IKLDA

Related Author

CHEN Guo-ming

YIN Jian

ZHOU Duan-ning

ZHANG Dong

Related Institution

School of Information Science and Technology, Sun Yat-sen University

College of Business,Eastern Washington University

School of Information Science and TechnologySun Yat-sen UniversityGuangzhouGuangdong 510275China

College of BusinessEastern Washington UniversitySpokane WA 99202USA

⁰