电子学报 ›› 2015, Vol. 43 ›› Issue (11): 2210-2217.DOI: 10.3969/j.issn.0372-2112.2015.11.011

• 学术论文 • 上一篇    下一篇

一种新的两分类器融合的空谱联合高光谱分类方法

孙乐1,2, 吴泽彬3, 冯灿4, 刘建军3, 肖亮3, 韦志辉3   

  1. 1. 江苏省网络监控工程中心, 江苏 南京 210044;
    2. 南京信息工程大学计算机与软件学院, 江苏 南京 210044;
    3. 南京理工大学计算机科学与工程学院, 江苏 南京 210094;
    4. 北方信息控制集团软件中心, 江苏 南京 211153
  • 收稿日期:2014-02-27 修回日期:2014-07-30 出版日期:2015-11-25
    • 作者简介:
    • 孙乐 男,1987年生于江苏宿迁.博士,南京信息工程大学讲师.研究方向为高光谱遥感图像解混、分类和目标识别.E-mail:sunlecncom@163.com;吴泽彬 男,1977年生于浙江杭州.南京理工大学计算机科学与工程学院副教授.研究方向为数据挖掘、虚拟仿真、高光谱图像处理.E-mail:zebin.wu@gmail.com
    • 基金资助:
    • 国家自然科学基金 (No.61101194,No.61301215,No.61301217); 国家自然科学基金面上项目 (No.61471199); 江苏省气象探测与信息处理重点实验室开放课题 (No.KDXS1404); 江苏省自然科学基金 (青年项目) (No.BK20150923); 南京信息工程大学人才启动经费; 江苏省光谱成像与智能感知重点实验室开放课题

A Novel Two-Classifier Fusion Method for Spectral-Spatial Hyperspectral Classification

SUN Le1,2, WU Ze-bin3, FENG Can4, LIU Jian-jun3, XIAO Liang3, WEI Zhi-hui3   

  1. 1. Jiangsu Engineering Center of Network Monitoring, Nanjing, Jiangsu 210044, China;
    2. School of Computer and Software, Nanjing University of Information Science and Technology, Nanjing, Jiangsu 210044, China;
    3. Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, Jiangsu 210094, China;
    4. Software Center of China North Industries Group Corporation, Nanjing, Jiangsu 211153, China
  • Received:2014-02-27 Revised:2014-07-30 Online:2015-11-25 Published:2015-11-25
    • Supported by:
    • National Natural Science Foundation of China (No.61101194, No.61301215, No.61301217); National Natural Science Foundation of China (No.61471199); Open Project of Jiangsu key Laboratory of Meteorological Observation and Information Processing of Nanjing Province (No.KDXS1404); Natural Science Foundation of Jiangsu Province,  China  (Youth Project) (No.BK20150923); Talents Research Fund of Nanjing University of Information Science and Technology; Open Program of Key Laboratory of Spectral Imaging and Intellisense of Jiangsu Province

摘要:

本文提出一种两分类器融合的高光谱空谱联合分类方法,首先利用子空间多项式逻辑回归在图像的特征子空间中分类,得到满概率图;根据满概率将每个像元分至概率最大的两个最可信类别,并在原始空间中构建最可信类别字典,利用稀疏解混对每个像元在最可信类别字典下进行稀疏表示,得到稀疏概率图;最后将满概率图和稀疏概率图线性融合,并利用边缘保持的马尔可夫正则项挖掘图像空间信息,得到具有边缘保持的空谱分类模型.实验表明,提出的两分类器融合方法即使在训练样本较少时也比现有方法得到更好的分类结果.

关键词: 高光谱分类, 子空间逻辑回归, 稀疏解混, 多分类器, 马尔可夫正则项

Abstract:

This paper presents a new multiple-classifier approach for spectral-spatial classification of hyperspectral images(HSI).Firstly,subspace based multinomial logistic regression(MLRsub) method is used to calculate the full probability of each pixel in the feature space;Secondly,the sub-dictionary is constructed by the training samples of the most two reliable classes,which is determined by the full probability for each pixel.Then,sparse unmixing(SU) is used to calculate the sparse probability in the original HSI.Finally,the full probability and sparse probability are fused linearly and the spatial information is exploit by an edge preserving Markov random field(MRF) regularizer.Experimental results indicate that our proposed multiple-classifier leads to better classification performance than the state-of-the-art methods,even with small training samples.

Key words: hyperspectral classification, subspace multinomial logistic regression, sparse unmixing, multiple classifier, MRF regularizer

中图分类号: