|
|
论文:2012,Vol:30,Issue(5):640-646 |
|
|
引用本文： |
|
|
梅少辉, 何明一, 戴玉超. 基于双向信号子空间投影的高光谱图像虚拟维数估计[J]. 西北工业大学 |
|
|
Mei Shaohui, He Mingyi, Dai Yuchao. A Novel DSSP (Double Signal Subspace Projection) Algorithm for Better VD (Virtual Dimensionality) Estimation[J]. Northwestern polytechnical university |
|
|
|
|
|
|
|
| 基于双向信号子空间投影的高光谱图像虚拟维数估计 |
|
| 梅少辉, 何明一, 戴玉超 |
|
| 西北工业大学 电子信息学院 陕西省信号与信息处理重点实验室, 陕西 西安 710072 |
| 摘要: |
| 提出一种基于双向信号子空间投影的高光谱图像虚拟维数估计算法。该算法分别在高光谱图像的像元方向和波段图像方向进行信号子空间估计,虽然这两个方向上信号子空间的分布不同,但其维数均等于图像的虚拟维数。该方法不需要对信号子空间和噪声子空间进行区分,仅通过对不同方向上的信号子空间投影进行比较,获取图像的虚拟维数。仿真像元实验和实际高光谱图像实验均证明该算法改善了传统的基于单向投影的虚拟维数估计算法的性能,其性能优于常用的虚拟维数估计算法:Neyamn-Pearson检测算法和信号子空间估计算法。 |
| 关键词:
虚拟维数
本证维数
高光谱图像
混合像元分解
|
|
| A Novel DSSP (Double Signal Subspace Projection) Algorithm for Better VD (Virtual Dimensionality) Estimation |
|
| Mei Shaohui, He Mingyi, Dai Yuchao |
|
| Department of Electronics Engineering,Northwestern Polytechnical University,Xi'an 710072,China |
| Abstract: |
| Sections 1 and 2 of the full paper explain and evaluate our DSSP algorithm as mentioned in the title; webelieve it is novel and its VD estimation is better.The core of sections 1 and 2 consists of: (1).the pixel repre-sentation and image representation of a hyperspectral image are utilized to generate respectively two sets of sub-spaces according to the principal component analysis (PCA); (2).when the dimensionality of these two sets ofsubspaces exceeds VD of the image, both subspace projections show the same reconstruction performance; there-fore, VD can be estimated by judging the difference between reconstruction performances of these two subspace pro-jections in DSSP instead of distinguishing signal subspace from noisy subspace; (3).the results of both syntheticand real hyperspectral experiments, given in Figs.1 through 9, demonstrate preliminarily that the performance oftraditional signal subspace projection based VD estimation has been improved and that the performance of the pro-posed DSSP based VD estimation algorithm outperforms those of the noise whitened HFC (NWHFC) and signalsubspace estimation (SSE) based VD estimation algorithms. |
| Key words:
algorithms
classification (of information)
data processing
eigenvalues and eigen functions
estima-tion
feature extraction
image processing
remote sensing
signal processing
singular value decom-position
spectrum analysis;DSSP (Double Signal Subspace Projection)
hyperspectral
spectralmixture analysis
VD (Virtual Dimensionality)
ID (intrinsic dimensionality)
|
|
| 收稿日期: 2011-10-23
修回日期:
|
| DOI: |
| 基金项目: 国家自然科学基金(61171154,61201324);教育部博士研究生学术新人奖;西北工业大学基础研究基金资助 |
| 通讯作者:
Email： |
| 作者简介: 梅少辉(1984-)，西北工业大学讲师，主要从事高光谱遥感图像处理研究。
|
|
|
|
|
|
|
|
参考文献: |
|
|
[1] Chang Chein-I,Du Q.Estimation of Number of Spectrally Distinct Signal Sources in Hyperspectral Imagery.IEEE Trans on Ge-oscience and Remote Sensing,2004,42(3): 608-619
[2] Wu Chaocheng,Liu Weimin,Chang Chein-I.Exploration of Methods for Estimation of Number of Endmembers in HyperspectralImagery.Proceedings of SPIE,2006,6378-6387
[3] Akaike H.A New Look at the Statistical Model Identification.IEEE Trans on Automatic Control,1974,AC-19: 716-723
[4] Rissanen J.Modeling by Shortest Data Description.Automatica,1978,14: 465-471
[5] Wu H T,Yang J F,Chen F K.Source Number Estimators Using Transformed Gerschgorin Radii.IEEE Trans on Signal Pro-cessing,1995,43(6): 1325-1333
[6] Dias Jos',Nascimento Jos'.Hyperspectral Subspace Identification.IEEE Trans on Geoscience and Remote Sensing,2008,46(8): 2435-2445
[7] Mei Shaohui,He Mingyi,et al.Improving Spatial-Spectral Endmember Extraction in the Presence of Anomalous Ground Ob-jects.IEEE Trans on Geoscience and Remote Sensing,2011,49(11): 4210-4222
[8] He Mingyi,Mei Shaohui.Dimension Reduction by Random Projection for Endmember Extraction.IEEE Conference on Industri-al Electronics and Applications,2010
[9] Clark R N,Swayze G A,et al.Imaging Spectroscopy: Earth and Planetary Remote Sensing with the USGS Tertracorder and Ex-pert Systems.Journal of Geophysical Research,2003,108: 5131
[10] Landgrebe D.Multispectral Data Analysis: A Signal Theory Perspective.Purdue University West Lafayette,1998,Technique Re-port |
|
|
|
|
|
|
|
