首页 > 论文 > 激光与光电子学进展 > 55卷 > 10期(pp:103002--1)

基于AP聚类的高光谱波段选择

Hyperspectral Band Selection Based on Affinity Propagation Clustering

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

摘要

波段选择在降维的同时能够保留高光谱数据的物理意义, 在很多方面具有一定应用。近邻传播(AP)算法根据数据点之间的相关性进行聚类, 将所有数据点视为潜在的聚类中心。提出了一种基于AP聚类的波段选择方法, 利用光谱信息散度和光谱相关角(SID-SCA)与光谱信息散度和光谱梯度角(SID-SGA)改进AP算法中相似度的计算。将降维结果输入支持向量机(SVM)分类器进行分类, 计算分类准确性, 并通过数据集Indiana Pines进行验证。实验结果表明:所提方法能够更好地提取波段的信息, 得到更高的分类精度。

Abstract

Band selection can preserve the physical meaning of hyperspectral data while reducing dimension, and has application in many aspects. The cluster of affinity propagation (AP) algorithm is according to the correlation of data points, and the AP algorithm regards all data points as potential clustering centers. We propose a band selection method based on AP clustering, which uses spectral information divergence and spectral correlation angle (SID-SCA), and spectral information divergence and spectral gradient angle (SID-SGA) to improve the similarity calculation in AP algorithm. The reducing dimension results are input into the support vector machine (SVM) classifier to classify, and the classification accuracy is calculated and verified using the data set Indiana Pines. The experimental results reveal that the proposed method can better extract the information of the band and obtain a high classification accuracy.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:O433.4

DOI:10.3788/lop55.103002

所属栏目:光谱学

基金项目:国家重点实验室基础研究项目

收稿日期:2018-03-03

修改稿日期:2018-04-23

网络出版日期:2018-04-28

作者单位    点击查看

任智伟:航天工程大学航天信息学院, 北京 101416
吴玲达:航天工程大学航天信息学院, 北京 101416

联系人作者:任智伟(juimer@foxmail.com)

【1】Hughes G. On the mean accuracy of statistical pattern recognizers[J]. IEEE Transactions on Information Theory, 1968, 14(1): 55-63.

【2】Zhang L P, Zhang L F. Hyperspectral remote sensing [M]. Beijing: Surveying and Mapping Press, 2011: 71-74.
张良培, 张立福. 高光谱遥感[M]. 北京: 测绘出版社, 2011: 71-74.

【3】Frey B J, Dueck D. Clustering by passing messages between data points[J]. Science, 2007, 315(5814): 972-976.

【4】Xia C M, Ni Z W, Ni L P, et al. Affinity propagation clustering algorithm based on density adjustment and manifold distance[J]. Computer Science, 2017, 44(10): 187-192, 215.
夏春梦, 倪志伟, 倪丽萍, 等. 基于密度调整和流形距离的近邻传播算法[J]. 计算机科学, 2017, 44(10): 187-192, 215.

【5】Chang C I. Spectral information divergence for hyperspectral image analysis[C]. IEEE 1999 International Geoscience and Remote Sensing Symposium, 1999: 6409333.

【6】Zhao C H, Tian M H, Li J W. Research progress on spectral similarity metrics[J]. Journal of Harbin Engineering University, 2017, 38(8): 1179-1189.
赵春晖, 田明华, 李佳伟. 光谱相似性度量方法研究进展[J]. 哈尔滨工程大学学报, 2017, 38(8): 1179-1189.

【7】Angelopoulou E, Lee S W, Bajcsy R. Spectral gradient: a material descriptor invariant to geometry and incident illumination[C]. Proceedings of the Seventh IEEE international Conference on Computer Vision, 1999: 861.

【8】Zhang X B, Yuan Y, Jing J J, et al. Spectral discrimination method information divergence combined with gradient angle[J]. Spectroscopy and Spectral Analysis, 2011, 31(3): 853-857.
张修宝, 袁艳, 景娟娟, 等. 信息散度与梯度角正切相结合的光谱区分方法[J]. 光谱学与光谱分析, 2011, 31(3): 853-857.

【9】Wu H, Xu Y J, Gao R. Extraction of alteration information from hyperspectral imagery based on SCA and SID[J]. Geography and Geo-Information Science, 2016, 32(1): 44-48.
吴浩, 徐元进, 高冉. 基于光谱相关角和光谱信息散度的高光谱蚀变信息提取[J]. 地理与地理信息科学, 2016, 32(1): 44-48.

【10】Liu X S, Ge L, Wang B, et al. An unsupervised band selection algorithm for hyperspectral imagery based on maximal information[J]. Journal of Infrared and Millimeter Waves, 2012, 31(2): 166-170, 176.
刘雪松, 葛亮, 王斌, 等. 基于最大信息量的高光谱遥感图像无监督波段选择方法[J]. 红外与毫米波学报, 2012, 31(2): 166-170, 176.

【11】Liu C H, Zhao C H, Zhang L Y. A new method of hyperspectral remote sensing image dimensional reduction[J]. Journal of Image and Graphics, 2005, 10(2): 218-222.
刘春红, 赵春晖, 张凌雁. 一种新的高光谱遥感图像降维方法[J]. 中国图象图形学报, 2005, 10(2): 218-222.

引用该论文

Ren Zhiwei,Wu Lingda. Hyperspectral Band Selection Based on Affinity Propagation Clustering[J]. Laser & Optoelectronics Progress, 2018, 55(10): 103002

任智伟,吴玲达. 基于AP聚类的高光谱波段选择[J]. 激光与光电子学进展, 2018, 55(10): 103002

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