混合PCA/ICA与JPEG2000结合的高光谱图像压缩

J4 ›› 2016, Vol. 38 ›› Issue (05): 968-974.

混合PCA/ICA与JPEG2000结合的高光谱图像压缩

叶珍1,白璘1,刘玉2,何明一3,粘永健2

（1.长安大学电子与控制工程学院，陕西西安 710064；2.第三军医大学生物医学工程系，重庆 400038；
3.西北工业大学电子信息学院，陕西西安 710129）

收稿日期:2015-11-20 修回日期:2016-01-06 出版日期:2016-05-25 发布日期:2016-05-25
基金资助:
国家自然科学基金(41201363,51407012)；中央高校基本科研业务费专项资金（310832163402,310832161001,310832161007）

Hyperspectral image compression using
mixed PCA/ICA in conjunction with JPEG2000

YE Zhen1,BAI Lin1,LIU Yu2,HE Mingyi3,NIAN Yongjian2

(1.School of Electronics and Control Engineering,Chang’an University,Xi’an 710064;
2.School of Biomedical Engineering,Third Military Medical University,Chongqing 400038;
3.School of Electronics and Information,Northwestern Polytechnical University,Xi’an 710129,China)

Received:2015-11-20 Revised:2016-01-06 Online:2016-05-25 Published:2016-05-25

摘要/Abstract

摘要：

主成分分析（PCA）常常结合JPEG2000压缩标准用来对高光谱图像进行压缩。然而，由PCA得到的主成分仅利用了二阶统计信息。对于高光谱图像应用来说，只采用二阶统计信息是远远不够的，如异常像素的处理常常需要用到更高阶的统计信息。研究了一种混合PCA/ICA与JPEG2000相结合的高光谱图像压缩算法。首先，对原始高光谱图像进行PCA变换，提取出前m个主成分对应的特征向量矩阵WPCA；然后，对其余的特征向量进行ICA变换，得到n个特征向量矩阵WICA；最后，将得到的混合投影矩阵、原始高光谱图像及其均值向量共同嵌入JPEG2000比特流，从而完成对高光谱图像的压缩。在不同码率的情况下，通过空间相关系数（ρ）、信噪比（SNR）、光谱角填图（SAM）等技术指标对混合PCA/ICA+JPEG2000算法的压缩性能进行评估。实验结果表明，混合PCA/ICA+JPEG2000算法不但能有效去除高光谱图像的谱间相关性，而且能够有效提高光谱保真度，保护异常像素信息。

关键词: 高光谱图像压缩, 主成分分析, 独立成分分析, JPEG2000

Abstract:

The principal component analysis (PCA) method combined with JPEG2000 is widely used in hyperspectral image compression. However, the covariance matrix of the PCA only represents the second order statistics. In many applications of hyperspectral image analysis, only preserving the information of the second order statistics is not sufficient. Taking the anomalous pixels for example, more subtle information needs to be captured by using higherorder statistics. To solve this problem, we propose a hyperspectral image compression algorithm using mixed PCA/ICA in conjunction with JPEG2000 standard. Firstly, the PCA is performed for the original hyperspectral image to find the eigenvector matrix WPCA corresponding to the first m largest eigenvalues. Then, the ICA is employed for the remaining eigenvectors to find n eigenvector matrix WICA. Finally, the mixed projection matrix, original hyperspectral image and it's mean vectors are embedded into the JPEG2000 bitstream for compression. At different bit rates, the performance of the mixed PCA/ICA+JPEG2000 is evaluated by spatial correlation coefficient (ρ), signal noise ratio (SNR) and spectral angel map (SAM). Experimental results reveal that the proposed algorithm is not only better in spectral correlation reduction, but also can improve spectral fidelity and protect anomaly pixel information.

Key words: hyperspectral image compression;principal component analysis (PCA);independent component analysis (ICA);JPEG2000

叶珍1,白璘1,刘玉2,何明一3,粘永健2. 混合PCA/ICA与JPEG2000结合的高光谱图像压缩[J]. J4, 2016, 38(05): 968-974.

YE Zhen1,BAI Lin1,LIU Yu2,HE Mingyi3,NIAN Yongjian2. Hyperspectral image compression using
mixed PCA/ICA in conjunction with JPEG2000 [J]. J4, 2016, 38(05): 968-974.

参考文献［12］

［1］	Wang Xianghai,Zhou Xia,Fang Lingling.Segmentation of hyperspectral remote sensing image using vector CV model［J］.Journal of Remote Sensing,2015,19(3):443450.（in Chinese）
［2］	Zhang Libao, Qiu Bingchang. Remote sensing image compression based on fast direction prediction［J］.Optics and Precision Engineering,2013,21(8):20952102.（in Chinese）
［3］	Wang Qicong,Wu Zebin,Liu Jianjun,et al.GPU based parrallel optimization of spatialspectral kernel sparse representation for hyperspectral image classification［J］.Computer Engineering & Science,2014,36(12):23212330.（in Chinese）
［4］	Xu Dawei,Zhang Rong,Wu Qian.Hyperspectral image compression using multiscale dictionary learning［J］.Journal of Remote Sensing,2015,19(2):263272.(in Chinese)
［5］	Liu G, Zhao F. Efficient compression algorithm for hyperspectral images based on correlation coefficient adaptive 3D zerotree coding［J］.IET Image Processing,2008,2(2):7282.
［6］	Ma Dongmei,Ma Caiwen.An algorithm of 3DSPECK for LASIS’s hyperspectral image compression［J］.Acta photonica sinica,2010,39(7):12481252.（in Chinese）
［7］	Licciardi G A,Chanussot J,Piscini A.Spectral compression of hyperspectral images by means of nonlinear principal component analysis decorrelation［C］∥ Proc of IEEE International Conference on Image Processing (ICIP),2014:50925096.
［8］	Barret M,Gutzwiller J L,Hariti M.Lowcomplexity hyperspectral image coding using exogenous orthogonal optimal spectral transform (orthost) and degree2 zerotrees［J］.IEEE Transactions on Geoscience and Remote Sensing,2011,49(5):15571566.
［9］	Du Q, Fowler J E. Hyperspectral image compression using JPEG2000 and principal component analysis［J］.IEEE Geoscience and Remote Sensing Letters,2007,4(2):201205.
［10］	Wang J,Chang C I.Mixed PCA/ICA spectral/spatial compression for hyperspectral imagery［C］∥Proc of Society of PhotoOptical Instrumentation Engineers(SPIE),2005:112.
［11］	Nian Yongjian,Wang Zhan,Wan Jianwei,et al.Compression technique for hyperspectral imagery oriented anomaly detection［J］.Journal of National University of Defense Technology,2009,31(3):4852.（in Chinese）
［12］	Liang Liang,Yang Minhua,Li Yingfang.Hyperspectral remote sensing image classification based on ICA and SVM algorithm［J］.Spectroscopy and Spectral Analysis,2010,30(10):27242728.（in Chinese）
［13］	Hyvarinen A,Oja E.A fast fixedpoint algorithm for independent component analysis［J］.Neural Computation,1997,9(7):14831492.
［14］	Molero J M,Garzon E M,Garcia I,et al.Analysis and optimizations of global and local versions of the RX algorithm for anomaly detection in hyperspectral data［J］.IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing,2013,6(2):801814.
［15］	GarciaAllende P B,Conde O M,Mirapeix J,et al.Data processing method applying principal component analysis and spectral angle mapper for imaging spectroscopic sensors［J］.Sensors Journal,2008,8(7):13101316.
	附中文参考文献:
［1］	王相海，周夏，方玲玲.矢量CV模型的高光谱遥感影像分割［J］.遥感学报，2015，19(3):443450.
［2］	张立保，丘兵昌.基于快速方向预测的高分辨率遥感影像压缩［J］.光学精密工程，2013，21(8):20952102.
［3］	王启聪，吴泽彬，刘建军，等.基于GPU的空谱联合核稀疏表示高光谱分类并行优化［J］.计算机工程与科学，2014，36(12):23212330.
［4］	徐大卫，张荣，吴倩.多尺度字典学习的高光谱图像压缩算法［J］.遥感学报，2015，19(2):263272.
［6］	马冬梅，马彩文.LASIS高光谱图像的3DSPECK压缩算法［J］.光子学报，2010，39(7):12481252.
［11］	粘永健，王展，万建伟，等.面向异常检测的高光谱图像压缩技术［J］.国防科技大学学报，2009，31(3):4852.
［12］	梁亮,杨敏华,李英芳.基于ICA与SVM算法的高光谱遥感影像分类［J］.光谱学与光谱分析，2010，30(10):27242728.

[1]	严春满, 张昱瑶, 张迪. 改进标签一致KSVD字典学习的人脸识别算法[J]. 计算机工程与科学, 2022, 44(02): 291-297.
[2]	党小超, 汝春瑞, 郝占军, . 基于CSI与SVM回归的室内定位方法[J]. 计算机工程与科学, 2021, 43(05): 853-861.
[3]	冯璋，裴东,王维. 基于改进灰狼算法优化支持向量机的人脸识别[J]. 计算机工程与科学, 2019, 41(06): 1057-1063.
[4]	火元莲，曹鹏飞，董俊松，石明. 基于融合特征的多尺度快速相关滤波跟踪算法[J]. 计算机工程与科学, 2019, 41(03): 559-566.
[5]	郭伟，董宏亮，赵德冀. 基于截断奇异值低秩矩阵恢复的Canny边缘检测算法[J]. 计算机工程与科学, 2018, 40(09): 1670-1678.
[6]	张金玲1，吕蕾2. 基于主成分分析和对数几率回归的硬件木马检测[J]. 计算机工程与科学, 2018, 40(07): 1187-1191.
[7]	张大明，张学勇，李璐，刘华勇. 图像矩阵上的广义最大噪声分离算法[J]. 计算机工程与科学, 2018, 40(05): 849-855.
[8]	蔡武越1,王珂2，郝玉洁2，段晓冉2. 一种Hadoop集群下的行为异常检测方法[J]. 计算机工程与科学, 2017, 39(12): 2185-2191.
[9]	李文轩1，李伟2，李梦凡1，刘成用1. 基于独立成分分析的优选N200和P300特征通道算法[J]. 计算机工程与科学, 2017, 39(09): 1682-1690.
[10]	林克正，魏颖，钟岩，李慧. 基于测地距离的KPCA人脸识别[J]. 计算机工程与科学, 2016, 38(09): 1858-1862.
[11]	辛月兰1,2，张晓华3，汪西莉1. 图割与非线性统计形状先验的图像分割[J]. J4, 2015, 37(03): 566-575.
[12]	罗晓峰，周建涛. 一种基于CRT与PCA的数字水印算法[J]. J4, 2014, 36(10): 1985-1990.
[13]	陈曦，梁方，王威. 基于对偶树复小波变换与PCA方法结合的图像变化检测算法研究[J]. J4, 2014, 36(08): 1560-1565.
[14]	李勇，刘云秀. 基于2DPCA-NSCT变换的多光谱与全色图像融合[J]. J4, 2013, 35(7): 143-148.
[15]	游鸿1,2，于文震1. 基于阵列的盲信号处理研究和应用[J]. J4, 2012, 34(4): 128-132.