电子学报 ›› 2020, Vol. 48 ›› Issue (6): 1099-1107.DOI: 10.3969/j.issn.0372-2112.2020.06.008

• 学术论文 • 上一篇    下一篇

面向高光谱遥感影像分类的监督多流形鉴别嵌入方法

黄鸿, 王丽华, 石光耀   

  1. 重庆大学光电技术及系统教育部重点实验室, 重庆 400044
  • 收稿日期:2019-01-22 修回日期:2019-05-13 出版日期:2020-06-25 发布日期:2020-06-25
  • 作者简介:黄 鸿 男,1980年生于湖南新宁,2008年获重庆大学博士学位,现为重庆大学教授、博士生导师.主要研究方向为无人机遥感、遥感信息处理、流形学习、稀疏表示等. E-mail:hhuang@cqu.edu.cn
    王丽华 女,1993年生于云南宣威,硕士研究生,2017年获郑州大学学士学位,主要研究方向为图像处理、遥感影像分类等. E-mail:20170802019t@cqu.edu.cn
    石光耀 男,1988年生于河南项城,博士研究生, 2015、2017年于重庆大学分别获得学士和硕士学位,主要从事图像处理、遥感影像分类、机器视觉等方面的研究. E-mail:shiguangyao@cqu.edu.cn
  • 基金资助:
    重庆市基础研究与前沿探索项目(No.cstc2018jcyjAX0093);国家自然科学基金(No.41371338);重庆市研究生科研创新项目(No.CYB19039)

Supervised Multi-manifold Discriminant Embedding Method for Hyperspectral Remote Sensing Image Classification

HUANG Hong, WANG Li-hua, SHI Guang-yao   

  1. Key Laboratory of Optoelectronic Technique System of the Ministry of Education, Chongqing University, Chongqing 400044, China
  • Received:2019-01-22 Revised:2019-05-13 Online:2020-06-25 Published:2020-06-25

摘要: 流形学习方法可以发现嵌入于高维观测数据中的低维流形结构,但是传统的流形学习算法都是假设所有数据位于单一流形上,忽略了高维数据中不同的子集可能存在不同的流形.针对上述问题,本文提出一种监督多流形鉴别嵌入的维数约简方法,并应用于高光谱遥感影像分类.该方法首先利用样本数据的类别标签进行多子流形划分,在此基础上采用图嵌入理论构造流形内图和流形间图,然后通过最小化流形内距离同时最大化流形间距离以增强类内数据聚集性和类间数据分散性,提取低维鉴别特征,改善地物分类性能.在University of Pavia (PaviaU)和Kennedy Space Center (KSC)高光谱数据集上的实验表明,相较于其他单流形算法和多流形算法,该方法取得了更高的分类精度,在随机选取2%训练样本时,其总体分类精度分别达到88.04%和84.53%,有效提升了地物分类性能.

关键词: 高光谱遥感影像, 分类, 特征提取, 图嵌入, 多流形学习

Abstract: Manifold learning method can find the low-dimensional manifold structures embedded in high-dimensional data.However,the traditional manifold learning algorithms assume that all samples lie on a single manifold,while the samples in different subsets may belong to different sub-manifolds.To solve the above problem,a new dimensionality reduction (DR) method termed supervised multi-manifold discriminant embedding (SMMDE) is proposed for classification of hyperspectral remote sensing image.At first,the proposed method explore the labels of HSI data to divide samples into different sub-manifolds.Based on the graph embedding framework,the intra-manifold and inter-manifold graphs are constructed to represent the multi-manifold structure of HSI data,and the intra-class aggregation and inter-class separation are enhanced by minimizing the intra-manifold distance and maximizing the inter-manifold distance simultaneously.Therefore,low-dimensional discriminant features are obtained to improve the performance of HSI classification.Experimental results on the PaviaU and KSC hyperspectral data sets show that the overall classification accuracies respectively reach 88.04% and 84.53% when 2% training samples are randomly selected for training.The proposed SMMDE method can effectively improve classification performance compared with many state-of-art DR algorithms.

Key words: hyperspectral remote sensing image, classification, feature extraction, graph embedding, multi-manifold learning

中图分类号: