首页 > 论文 > 光学 精密工程 > 28卷 > 2期(pp:443-456)

半监督多图嵌入的高光谱影像特征提取

Feature extraction of hyperspectral image with semi-supervised multi-graph embedding

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

针对传统图嵌入方法仅采用单一图结构无法有效表征高维数据中复杂本征结构, 本文提出了一种半监督多图嵌入(SSMGE)方法, 并应用于高光谱影像特征提取。该方法首先利用标记样本的类内、类间近邻点来构建类内超图、类间超图、类内普通图、类间普通图, 然后通过无标记样本的近邻点和远离点构建无监督本征超图和惩罚超图, 并以多图协同方式来表征高维数据间的复杂几何关系, 实现鉴别特征提取。本文提出的SSMGE方法不仅能有效揭示数据点间超图和普通图的结构, 而且在低维嵌入空间中增强同类数据聚集性和非同类数据的远离性, 提取的鉴别特征可改善地物分类精度。在PaviaU和Urban高光谱数据集上进行了分类实验, 本文方法的总体分类精度分别可达到85.92%和79.74%。相比普通图嵌入和超图方法, 该算法明显提升了地物的分类性能。

Abstract

Traditional graph embedding methods often use single graph structures for feature extraction (FE).However, these methods cannot effectively represent the complex intrinsic structuresof high-dimensional data. To address this problem, a Semi-Supervised Multi-Graph Embedding (SSMGE) algorithm was proposed for FE of Hyper Spectral Images (HSIs). First, the SSMGE method constructed one of each of intra-and inter-class hypergraphs and intra-and inter-class graphs through intra-and inter-class neighbors of labeled samples.In addition, it constructs an unsupervised intrinsic hypergraph and a penalty hypergraph using unlabeled samples. The fusion of graphs and hypergraphs can effectively characterize the complex relationships in high-dimensional data. The SSMGE method not only effectively reveals the intrinsic structure of an HSI by exploring the collaboration of graphs and hypergraphs but also enhances the discriminative ability of extracted features in low-dimensional embedding space.This enables improved classification performance of HSI data. Experimental results on the PaviaU and Urban hyperspectral datasets show that the overall accuracies of the proposed method reached 85.92% and 79.74%, respectively. The SSMGE method can significantly improve classification performance as compared with some state-of-the-art FE methods.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:TP394.1;TH691.9

DOI:10.3788/ope.20202802.0443

所属栏目:信息科学

基金项目:重庆市基础研究与前沿探索项目资助(No.cstc2018jcyjAX0093); 国家自然科学基金资助(No.41371338); 重庆市研究生科研创新项目资助(No.CYB18048, No.CYS18035)

收稿日期:2019-05-07

修改稿日期:2019-07-06

网络出版日期:--

作者单位    点击查看

黄 鸿:重庆大学 光电技术与系统教育部重点实验室, 重庆 400044
唐玉枭:重庆大学 光电技术与系统教育部重点实验室, 重庆 400044
段宇乐:重庆大学 光电技术与系统教育部重点实验室, 重庆 400044

联系人作者:黄鸿(hhuang@cqu.edu.cn)

备注:黄 鸿(1980-), 男, 湖南新宁人, 教授, 博士生导师, 2003、2005、2008年于重庆大学分别获得学士、硕士和博士学位, 主要从事流形学习、模式识别、遥感影像智能化处理等方面的研究。

【1】闫敬文, 陈宏达, 刘蕾. 高光谱图像分类的研究进展[J]. 光学 精密工程, 2019, 27(3): 680-693.
YAN J W, CHEN H D, LIU L. Research progress in classification of hyperspectral images[J]. Opt. Precision Eng., 2019, 27(3): 680-693. (in Chinese)

【2】黄鸿, 陈美利, 段宇乐, 等. 空-谱协同流形重构的高光谱影像分类[J]. 光学 精密工程, 2018, 26(7): 1827-1836.
HUANG H, CHEN M L, DUAN Y L, et al.. Hyperspectral image classification with spatial-spectral collaboration and manifold reconstruction error [J]. Opt. Precision Eng., 2018, 26(7): 1827-1836. (in Chinese).

【3】CHEN M L, WANG Q, LI X L. Discriminant analysis with graph learning for hyperspectral image classification [J]. Remote Sensing, 2018, 10(6): 836.

【4】HE L, LI J, LIU C, et al.. Recent advances on spectral-spatial hyperspectral image classification: an overview and new guidelines[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 56(3): 1579-1597.

【5】王庆超, 付光远, 汪洪桥, 等. 多核融合多尺度特征的高光谱影像地物分类[J]. 光学 精密工程, 2018, 26(4): 980-988.
WANG Q CH, FU G Y, WANG H Q , et al.. Fusion of multi-scale feature using multiple kernel learning for hyperspectral image land cover classification[J]. Opt. Precision Eng., 2018, 26(4): 980-988. (in Chinese).

【6】李志敏, 张杰, 黄鸿, 等. 面向高光谱图像分类的半监督丛流形学习[J]. 光学 精密工程, 2015, 23(5): 1434-1442.
LI ZH M, ZHANG J, HUANG H, et al.. Semi-supervised bundle manifold learning for hyperspectral image classification [J]. Opt. Precision Eng., 2015, 23(5): 1434-1442. (in Chinese).

【7】MENEZES J, POOJARY N. Dimensionality reduction and classification of hyperspectral images using DWT and DCCF[C]. Mec International Conference on Big Data and Smart City. IEEE, 2016: 47-52.

【8】KANG X, XIANG X, LI S, et al.. PCA-based edge-preserving features for hyperspectral image classification[J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(12): 7140-7151.

【9】WEN J, FANG X, CUI J, et al.. Robust sparse linear discriminant analysis[J]. IEEE Transactions on Circuits & Systems for Video Technology, 2018, 29(2): 390-403.

【10】XU J C, MU H Y, WANG Y, et al.. Feature genes selection using supervised locally linear embedding and correlation coefficient for microarray classification [J]. Computational and Mathematical Methods in Medicine, 2018, 2018: 1-11.

【11】XU X, YANG H, LIAN C, et al.. Self-learning control using dual heuristic programming with global Laplacian Eigenmaps[J]. IEEE Transactions on Industrial Electronics, 2017, 64(12): 9517-9526.

【12】SUN W, YANG G, DU B, et al.. A sparse and low-rank near-isometric linear embedding method for feature extraction in hyperspectral imagery classification [J]. IEEE Transactions on Geoscience and Remote Sensing, 2017, 55(7): 4032-4046.

【13】YONG Y H, XIAO Q Z. Multi-phase batch process monitoring based on multiway weighted global neighborhood preserving embedding method[J]. Journal of Process Control, 2018, 69: 44-57.

【14】YU Q, WANG R, LI B N, et al.. Robust locality preserving projections with cosine-based dissimilarity for linear dimensionality reduction [J]. IEEE Access, 2017, 5(99): 2676-2684.

【15】ZHONG J, PANG Y W, YUAN Y, et al.. Relevance and irrelevance graph based marginal Fisher analysis for image search reranking[J]. Signal Processing, 2016, 121: 139-152.

【16】VAN M, KANG H J. Bearing defect classification based on individual wavelet local Fisher discriminant analysis with particle swarm optimization [J]. IEEE Transactions on Industrial Informatics, 2017, 12(1): 124-135.

【17】LUO F L, HUANG H, DUAN Y, et al.. Local geometric structure feature for dimensionality reduction of hyperspectral imagery[J]. Remote Sensing, 2017, 9(8): 790.

【18】DHAMECHA T I, SINGH R, VATASA M. On incremental semi-supervised discriminant analysis[J]. Pattern Recognition, 2016, 52: 135-147.

【19】GAO S, ZHOU J, YAN Y, et al.. Recursively global and local discriminant analysis for semi-supervised and unsupervised dimension reduction with image analysis[J]. Neurocomputing, 2016, 216: 672-683.

【20】Luo F L, HUANG H, MA Z, et al.. Semisupervised sparse manifold discriminative analysis for feature extraction of hyperspectral images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2016, 54(10): 6197-6211.

【21】YUAN H L, TANG Y Y. Learning with hypergraph for hyperspectral image feature extraction[J]. IEEE Geoscience & Remote Sensing Letters, 2015, 12(8): 1695-1699.

【22】HUANG S, YANG D, GE Y X, et al.. Discriminant hyper-Laplacian projections and its scalable extension for dimensionality reduction[J]. Neurocomputing, 2016, 173(2): 145-153.

【23】SUN Y, WANG S, LIU Q, et al.. Hypergraph embedding for spatial-spectral joint feature extraction in hyperspectral images[J]. Remote Sensing, 2017, 9(5): 506-.520

【24】WANG S, SUN Y, HANG R, et al.. Spatial-spectral locality constrained elastic net hypergraph for hyperspectral image clustering[J]. International Journal of Remote Sensing, 2017, 38(23): 7374-7388.

【25】YAN S, XU D, ZHANG B, et al.. Graph embedding and extensions: A general framework for dimensionality reduction[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2007, 29(1): 40-51.

引用该论文

HUANG Hong,TANG Yu-xiao,DUAN Yu-le. Feature extraction of hyperspectral image with semi-supervised multi-graph embedding[J]. Optics and Precision Engineering, 2020, 28(2): 443-456

黄 鸿,唐玉枭,段宇乐. 半监督多图嵌入的高光谱影像特征提取[J]. 光学 精密工程, 2020, 28(2): 443-456

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF