Please wait a minute...
img

官方微信

遥感技术与应用  2009, Vol. 24 Issue (3): 385-390    DOI: 10.11873/j.issn.1004-0323.2009.3.385
技术研究与图像处理     
基于快速非负矩阵分解和RBF网络的高光谱图像分类算法
狄文羽,何明一,梅少辉
(陕西省信息获取与处理重点实验室,西北工业大学电子信息学院,陕西 西安〓710072)
A Hyper-spectral Image Classification Algorithms Based on QuickNon-negative Matrix Factorization and RBF Neural Network
DI Wen-yu,HE Ming-yi,MEI Shao-hui
(School of Electronics and Information|Northwestern Polytechnical University| Xi'an 710072 China)
 全文: PDF(1109 KB)  
摘要:

提出一种处理AVIRIS高光谱图像数据的计算机分类算法。首先采用投影梯度(Projected Gradient)改进的非负矩阵分解(NMF)方法对高光谱数据进行特征提取,大大降低了分解过程中两个子迭代问题的时间复杂度,而后利用径向基函数神经网络(RBFNN)分类器对提取结果进行分类。结果表明,与传统NMF和主成分分析相比,PGNMF\|RBF算法消耗时间最少,分类精度最高,6类地物的分类精度达到83.34%。该算法在保留非负矩阵分解明确物理意义的基础上,获得了更快的分解速度和更高的分类精度,在高光谱图像分类领域具有较大的应用潜力。

关键词: 投影梯度非负矩阵分解RBF神经网络图像分类    
Abstract:

A new method combined Non-negative Matrix Factorization (NMF) with Projected Gradient (PG) is proposed for hyper-spectral image classification.Projected Gradient method demonstrates much faster convergence than the popular multiplicative update approach in the iteration process of two sub-problems from NMF thus effectively maintains higher classification accuracy than traditional methods; RBF neural network achieves higher accuracy and faster classification process compared to BP network.The new method combines the advantages of the above two,applying PGNMF for feature extraction and RBFNN as classifier.The experiment shows that compared to traditional NMF and PCA,PGNMF-RBF has higher accuracy for classification and less time consumption.The classification accuracy for 6 classes reaches 83.34%.This paper demonstrates PGNMF-RBF an effective and promising method in hyper-spectral image classification.

Key words: Projected gradient    Non-negative matrix factorization    RBF neural network    Image classification
收稿日期: 2008-12-27 出版日期: 2010-01-20
:  TP 391  
作者简介: 狄文羽(1984-)|女|硕士研究生|主要从事遥感图像处理与分析研究。E-mail:diwenyu@gmail.com。
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
RSS
作者相关文章  
狄文羽
何明一
梅少辉

引用本文:

狄文羽,何明一,梅少辉. 基于快速非负矩阵分解和RBF网络的高光谱图像分类算法[J]. 遥感技术与应用, 2009, 24(3): 385-390.

DI Wen-yu,HE Ming-yi,MEI Shao-hui. A Hyper-spectral Image Classification Algorithms Based on QuickNon-negative Matrix Factorization and RBF Neural Network. Remote Sensing Technology and Application, 2009, 24(3): 385-390.

链接本文:

http://www.rsta.ac.cn/CN/10.11873/j.issn.1004-0323.2009.3.385        http://www.rsta.ac.cn/CN/Y2009/V24/I3/385

[1] Lee C,Langgrebe D A.Feature Extraction and Classification Algorithms for High Dimensional Data[D].Purdue University:Phd Dissertation in School of Electrical Engineering,West Lafayette,Indiana,1992.


[2] Comon P.Independent Component Analysis:a New Concept,Signal Processing[J].Elsevier,1994,36(3):287-314.


[3] Chen Z,Li G.Robust Principal Components and Dispersion Matrices via Projection Pursuit[R].Dept.of Statistics,Harvard University,Report PJH-8,1981.


[4] Lee C,Langgrebe D A.Feature Extraction Based on Decision Boundaries[J].IEEE Transaction on Pattern Analysis and Machine Intelligence,1993,15:388-400.


[5] Li Le,Zhang Yujin.A Survey on Algorithms of Non-Negative Matrix Factorization[J].Acta Electronica Sinica,2008,36(4):737-743.[李乐,章毓晋.非负矩阵算法综述[J].电子学报,2008,36(4):737-743.]


[6] Lee D D,Seung H S.Learning the Parts of Objects with Non-negative Matrix Factorization[J].Nature,1999,401:788-791.


[7] Heiler M,Schnorr C.Learning Sparse Representations by Non-negative Matrix Factorization and Sequential Cone Programming[J].Journal of Machine Learning Research,2000,5(2):95-99.


[8] Zhang Jiang,Wang Nian,Liang Dong,et al.Image Classification Based on Non-negative Matrix Factorization and Adjacency Spectra[J].Journal of University of Science and Technology of China,2008,38(3):247-251.[张江,王年,梁栋,等.基于非负矩阵分解与邻接谱的图像分类[J].中国科学技术大学学报,2008,38(3):247-251.]


[9] Ren J M,Yu X C,Hao B X.Classification of Landsat TM Image Based on Non-Negative Matrix Factorization[C].IGARSS 2007.July 23-28,2007:405-408.


[10] Lee D D,Seung H S.Algorithms for Nonnegative Matrix Factorization[C].Adv in Neur.Inform.Proc.Syst.Cambridge:MIT Press,2000:556-562.


[11] Liu Weixiang,Zheng Nanning,You Qubo.Non-negative Matrix Factorization and Its Application in Pattern Recognition[J].Chinese Science Bulletin,2006,51(3):241-250.[刘维湘,郑南宁,游屈波.非负矩阵分解及其在模式识别中的应用[J].科学通报,2006,51(3):241-250.]


[12] Lin C J.Projected Gradient Methods for Non-negative Matrix Factorization[J].Neural Computation,2007,19(10):2756-2779.


[13] Jensen J R.Introductory Digital Image Processing:A Remote Sensing Perspective[M].Beijing:China Machine Press,2007.[Jensen J R著,陈晓玲等译.遥感数字影像处理导论[M].北京:机械工业出版社,2007.]


[14] Luo Jiancheng,Zhou Chenghu,Yang Yan.Radial Basis Function Map Theory Based Remote Sensing Image Classification Model[J].Journal of Image and Graphics,2003,12(3):7-9.[骆剑承,周成虎,杨艳.基于径向基函数(RBF)映射理论的遥感影像分类模型研究.中国图象图形学报,2003,12(3):7-9.]


[15] Langgrebe D A.Multispectral Data Analysis:A Signal Theory Perspective[EB/OL].http://cobweb.ecn.purdue.edu/~biehl/MultiSpec/documentation.html.


[16] Gu Yanfeng,Zhang Ye.Feature Extraction Based on Automatic Subspace Partition for Hyperspectral Images[J].Remote Sensing Technology and Application,2003,18(6):284-387.[谷延峰,张晔.基于自动子空间划分的高光谱数据特征提取[J].遥感技术与应用,2003,18(6):284-387.]

[1] 陈伟民,张凌,宋冬梅,王斌,丁亚雄,许明明,崔建勇. 基于AdaBoost改进随机森林的高光谱图像地物分类方法研究[J]. 遥感技术与应用, 2018, 33(4): 612-620.
[2] 郭宇柏,卓莉,陶海燕,曹晶晶,王芳. 基于空谱初始化的非负矩阵光谱混合像元盲分解[J]. 遥感技术与应用, 2018, 33(2): 216-226.
[3] 秦俊,冷寒冰,赵广琦,景军,周坚华. 物候和波谱—位置分析在城镇绿化植物群分类中的应用[J]. 遥感技术与应用, 2017, 32(5): 948-957.
[4] 苏红军,赵波. 基于共形几何代数的高光谱遥感波段选择方法[J]. 遥感技术与应用, 2017, 32(3): 539-545.
[5] 袁春琦,徐佳,程圆娥,陈媛媛,许康. 基于协同训练与集成学习的极化SAR图像半监督分类[J]. 遥感技术与应用, 2017, 32(2): 380-385.
[6] 崔艳荣,何彬彬,张瑛,李蔓. 非负矩阵分解融合高光谱和多光谱数据[J]. 遥感技术与应用, 2015, 30(1): 82-91.
[7] 李茜楠,苏红军. 基于萤火虫算法的高光谱遥感波段选择方法[J]. 遥感技术与应用, 2014, 29(5): 761-770.
[8] 袁宗泽,孙浩,计科峰,邹焕新. 基于Fisher字典学习稀疏表示的高光谱图像分类[J]. 遥感技术与应用, 2014, 29(4): 646-652.
[9] 李光辉,王 成,郑照军,习晓环,岳彩荣. 机载高光谱数据冰川分类方法研究——以“中习一号”冰川为例[J]. 遥感技术与应用, 2013, 28(5): 766-772.
[10] 曹晶晶,卓莉,王芳,陶海燕. 盲信号分离技术在高光谱混合像元分解中的应用[J]. 遥感技术与应用, 2013, 28(3): 488-495.
[11] 徐丽华,岳文泽,李先华,梅安新,章皖秋. 基于二维小波变换的遥感分类研究[J]. 遥感技术与应用, 2003, 18(5): 317-321.
[12] 赵书河 张明金. 带竞争学习机制的反向传播模型及在遥感图像分类中的实验[J]. 遥感技术与应用, 1995, 10(4): 61-64.