Independent component feature separation based on spatial down sample for hyperspectral image

ZHU Yuan-yuan; GAO Jiao-bo; GAO Ze-dong; FAN Zhe; YAN Shao-qi

doi:10.3788/OPE.20152311.3246

您当前的位置：

首页 >

文章列表页 >

Independent component feature separation based on spatial down sample for hyperspectral image

更新时间：2020-08-13

- Independent component feature separation based on spatial down sample for hyperspectral image
- Optics and Precision Engineering Vol. 23, Issue 11, Pages: 3246-3258(2015)
- 作者机构：
  
  西安应用光学研究所,陕西西安,710065
- 作者简介：
- 基金信息：
- DOI：10.3788/OPE.20152311.3246
  CLC： TP751.1
- Received：21 August 2015，
  
  Revised：29 September 2015，
  
  Published：25 November 2015
- 稿件说明：
移动端阅览
朱院院, 高教波, 高泽东等. 高光谱空间降采样独立成分特征分离[J]. 光学精密工程, 2015,23(11): 3246-3258

ZHU Yuan-yuan, GAO Jiao-bo, GAO Ze-dong etc. Independent component feature separation based on spatial down sample for hyperspectral image[J]. Editorial Office of Optics and Precision Engineering, 2015,23(11): 3246-3258
朱院院, 高教波, 高泽东等. 高光谱空间降采样独立成分特征分离[J]. 光学精密工程, 2015,23(11): 3246-3258 DOI： 10.3788/OPE.20152311.3246.

ZHU Yuan-yuan, GAO Jiao-bo, GAO Ze-dong etc. Independent component feature separation based on spatial down sample for hyperspectral image[J]. Editorial Office of Optics and Precision Engineering, 2015,23(11): 3246-3258 DOI： 10.3788/OPE.20152311.3246.

摘要

提出一种空间降采样独立成分特征分离方法

用于缩短独立成分分析(ICA)法在高光谱图像特征分离时的运算时间。该方法通过对高光谱图像的二维像素空间进行网格划分得到较小的窗口;基于光谱相似性测度法度量每个窗口中中心像素与周围像素的距离。然后舍弃这段距离小于阈值的周围像素

而将大于阈值的周围像素和中心像素作为样本量进行FastICA

获取投影矩阵变换原始数据

得到特征分离的ICA成分。对比了传统ICA与空间降采样(SDS)ICA(SDS_ICA)的性能

研究了降采样阈值参数、降采样窗口参数及初始投影矩阵对SDS_ICA特征分离性能及运行时间的影响。实验结果表明:应用SDS_ICA时

仅设置适中的阈值和不敏感的窗口大小参数

就能保持与传统ICA相近的特征分离性能

运行时间减少了30%以上。该方法适合应用于高光谱准实时特征提取、数据降维及目标探测等领域。

Abstract

A novel independent component feature separation based on Spatial Down Sample(SDS) was presented for solving the long run-time defection of traditional Independent Component Analysis(ICA). Small windows were obtained by gridding the two-dimensional spatial space of a hyperspectral image. In each window

the distance between the central pixel and around pixels was measured by spectral similarity and the around pixels whose distances were smaller than the threshold value were discarded. The projection matrix was calculated by FastICA with the central and the around pixels whose distances were larger than the threshold value. The feature separation ICA components were achieved by projecting the original hyperspectral image using a project matrix. The performance of traditional ICA and SDS_ICA were compared. The influences of threshold values

window size values and the initial projecting matrix on the feature separation performance and run-time of SDS_ICA were studied. Experiment results show that SDS_ICA has the similar feature separation performance with the traditional ICA and its run-time has reduced above 30% under moderate threshold values and insensitivity window sizes. The novel method can be widely applied in the fields of hyperspectral feature extraction

data reduction

target detection

etc

关键词

Keywords

references

HYVÄRINEN A, KARHUNEN J, OJA E.Independent Component Analysis[M]. New York:John Wiley & Sons, 2001.

WANG J, CHANG CH I. Applications of independent component analysis in endmember extraction and abundance quantification for hyperspectral imagery[J].IEEE Transactions on Geoscience and Remote Sensing, 2006,44(9):2601-2616.

CHANG CH I, JIAO X L, WU CH CH,et al.. Component analysis-based unsupervised linear spectral mixture analysis for hyperspectral imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(11):4123-4137.

WANG J,CHANG CH I. Independent component analysis-based dimensionality reduction with applications in hyperspectral image analysis[J].IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(6):1586-1600.

CHANG CH I.Hyperspectral Data Processing Algorithm Design and Analysis[M]. New Jersey:John Wiley & Sons, 2013.

杜鹏,赵慧洁. 基于抗噪声ICA的高光谱数据特征提取方法[J]. 北京航空航天大学学报, 2005, 31(10):1101-1105. DU P,ZHAO H J.Noise robust ICA feature extraction algorithm for hyperspectral image[J].Journal of Beijing University of Aeronautics and Astronautics, 2005, 31(10):1101-1105.(in Chinese)

付小宁, 张涛, 万里. 基于多光谱分离的烟雾检测[J]. 光学精密工程, 2013, 21(11):2798-2802. FU X N, ZHANG T, WAN L. Smoke detection based on multispectral separation[J].Opt. Precision Eng., 2013, 21(11):2798-2802.(in Chinese)

秦玉华, 丁香乾, 宫会丽. 高维特征选择方法在近红外光谱分类中的应用[J]. 红外与激光工程, 2013, 42(5):1355-1359. QIN Y H, DING X Q, GONG H L. High dimensional feature selection in near infrared spectroscopy classification[J].Infrared and Laser Engineering, 2013, 42(5):1355-1359.(in Chinese)

刘志刚, 卢云龙, 魏一苇. 有监督的高光谱图像伪装目标检测方法[J]. 红外与激光工程, 2013, 42(11):3076-3081. LIU ZH G, LU Y L, WEI Y W. Supervised method for hyperspectral image camouflage target detection[J].Infrared and Laser Engineering, 2013, 42(11):3076-3081.(in Chinese)

李平, 魏仲慧, 何昕,等. 采用多形状特征融合的多视点目标识别[J]. 光学精密工程,2014,22(12):3368-3376. LI P, WEI ZH H, HE X,et al.. Object recognition based on shape feature fusion under multi-views[J].Opt. Precision Eng.,2014,22(12):3368-3376.(in Chinese)

曾生根, 朱宁波, 包晔,等. 一种改进的快速独立分量分析算法及其在图象分离中的应用[J]. 中国图象图形学报, 2003, 8(10):1159-1165.(in Chinese) ZHENG SH G, ZHU N B, BAO Y,et at.. A modified fast independent component analysis and its application to image separation[J].Journal of Image and Graphics, 2003, 8(10):1159-1165.(in Chinese)

黄丽妍, 高强, 亢海燕,等. 改进的快速独立分量分析算法[J]. 华北电力大学学报, 2006, 33(3):59-62. HUANG L Y, GAO Q, KANG H Y,et al.. Improved fast ICA algorithm[J].Journal of North China Electric Power University, 2006, 33(3):59-62.(in Chinese)

季策, 于洋, 于鹏. 改进的独立分量分析算法[J]. 东北大学学报:自然科学版, 2010, 8(8):1086-1088. JI C, YU Y, YU P. Improved algorithm for independent component analysis[J].Journal of Northeastern University(Natural Science), 2010, 8(8):1086-1088.(in Chinese)

杨俊安, 庄镇泉, 吴波,等. 一种基于负熵最大化的改进的独立分量分析快速算法[J]. 电路与系统学报, 2002, 7(4):37-40. YANG J A, ZHUANG ZH Q, WU B,et al.. An improved fast ICA algorithm based on negentropy maximization[J]. 2002, 7(4):37-40.(in Chinese)

季策, 胡祥楠, 朱丽春,等. 改进的高阶收敛FastICA算法[J]. 东北大学学报:自然科学版, 2011, 32(10):1390-1393. JI C, HU X N, ZHU L CH,et al.. Improved higher order convergent fast ICA algorithm[J].Journal of Northeastern University(Natural Science), 2011, 32(10):1390-1393.(in Chinese)

季策, 王艳茹, 沙明博,等. 引入松弛因子的高阶收敛FastICA算法[J]. 东北大学学报:自然科学版, 2014, 35(2):204-207. JI C, WANG Y R, SHA M B,et al.. Relaxation factor-based fast ICA with higher order convergence[J].Journal of Northeastern University(Natural Science), 2014, 35(2):204-207.(in Chinese)

GU Y F,WANG CH,WANG SH ZH,et al.. Kernel-based regularized-angle spectral matching for target detection in hyperspectral imagery[J].Pattern Recognition Letters,2011,32:114-119.

王忠良,冯燕,肖华,等. 高光谱图像的分布式压缩感知成像与重构[J]. 光学精密工程, 2015, 23(4):1131-1137. WANG ZH L, FENG Y, XIAO H,et al.. Distributed compressive sensing imaging and reconstruction of hyperspectral imagery[J].Opt. Precision Eng., 2015, 23(4):1131-1137.(in Chinese)

殷亚男, 王晓东, 李丙玉. 基于预测和JPEG2000的MODIS红外辐射多光谱图像无损压缩算法[J]. 液晶与显示, 2013, 28(6):922-926. YIN Y N, WANG X D, LI B Y. Lossless compression method based on prediction and JPEG2000 for MODIS emissive IR bands multispectral image[J].Chinese Journal of Liquid Crystals and Displays,2013, 28(6):922-926.(in Chinese)

张茗璇, 高教波, 孟合民,等. 基于傅里叶变换光谱技术的Zoom-FFT算法研究[J]. 应用光学,2013,34(3):452-456. ZHANG M X, GAO J B, MENG H M,et al.. Zoom-FFT based on fourier transform spectroscopy[J].Journal of Applied Optics, 2013, 34(3):452-456.(in Chinese)

李宇, 高教波, 孟合民,等. 基于统一计算设备架构的干涉成像光谱快速反演技术研究[J]. 应用光学,2014,35(3):414-419. LI Y, GAO J B, MENG H M,et al.. Fast inversion techniques of inteferogram imaging spectrum base on CUDA[J].Journal of Applied Optics, 2014, 35(3):414-419.(in Chinese)

Views

597

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Specral-spatial classification of hyperspectral imagery with hybrid architecture of 3D-CNN and Transformer

Collaborative classification of hyperspectral and LiDAR data based on CNN-transformer

Hyperspectral unmixing with shared endmember variability in homogeneous region

Target detection using spectral unmixing

Partial optimal transport-based domain adaptation for hyperspectral image classification

Related Author

ZHANG Haokui

TAO Lijie

JING Haizhao

YUJI Iwahori

WU Haibin

DAI Shiyu

WANG Aili

YU Xiaoyu

Related Institution

Northwestern Polytechnical University

College of Electron and Information， University of Electronic Science and Technology of China，Zhongshan Institute

Department of Computer Science， Chubu University

Heilongjiang Province Key Laboratory of Laser Spectroscopy Technology and Application， College of Measurement and Control Technology and Communication Engineering， Harbin University of Science and Technology

School of Electronic Engineering， Xidian University

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China Postal code：130033
Tel：0431-86176855 Email：gxjmgc@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰