压缩感知和万有引力模型相结合的高光谱混合像元分解
Hyperspectral Pixel Unmixing Combined with the Compressive Sensing and the Universal Gravitation Model
Received date: 2013-12-03
Revised date: 2014-06-24
Online published: 2014-10-24
高光谱影像虽然具有较高的光谱分辨率，但因其空间分辨率低而普遍存在混合像元。混合像元分解是高光谱遥感应用的重要研究内容，包括端元提
取和端元丰度反演两部分。本文以压缩感知（Compressive Sensing，CS）理论为基础，结合神经网络技术提出了一种新的端元提取模型——基于CS的
高光谱影像端元提取模型。同时，将经典的万有引力模型（Universal Gravitation Model，UGM）引入到端元丰度反演中，提出基于UGM的端元丰度反
演算法。最后，以Hyperion高光谱影像为实验数据在MATLAB中对模型和算法进行编程实现，并对其结果进行精度评定。实验结果表明，本文提出的
提取模型与反演算法无论在理论上还是在实际操作中，都具有一定的可行性，同时精度也满足要求。
杨可明 王林伟 刘士文 刘飞 史钢强 赵思亮 . 压缩感知和万有引力模型相结合的高光谱混合像元分解[J]. 测绘学报, 2014 , 43(10) : 1068 -1074 . DOI: 10.13485/j.cnki.11-2089.2014.0171
Hyperspectral imagery has the characteristic of high spectral resolution, but the low spatial resolution makes the mixed pixels exist ubiquitously in them. Pixel unmixing is the
?important content in the field of hyperspectral remote sensing application, including abundance extracting and abundance inversing. Based on the compressive sensing (CS)
theory, combined with neural network technology, a novel hyperspectral endmember extracting model based on the compressive sensing theory is proposed. After that,
applied the classic Universal Gravitation Model (UGM) into abundance inversing, an abundance inversing algorithm based on the universal gravitation model is put forward.
Finally, the model and the algorithm are realized in MATLAB with the Hyperion hyperspectral image and the accuracy of the endmember is assessed and analyzed
according to the results. Experimental results show that the proposed extracting model and inversing algorithm have a certain degree of feasibility in both theory and practice,
at the same time the computational accuracy meets the requirements.
[1]LI Er-sen, ZHU Shu-long, ZHOU Xiao-ming, et al.The development and comparison of endmember extraction algorithms using hyperspectral imagery[J].Journal of Remote Sensing, 2011, 15(4):659-679
[2]Plaza A, Chang C I.An improved N-FINDR algorithm in implementation[J].SPIE Proceedings of Algorithms and Technologies for Multispectral, Hyperspectral and Untraspectral Imagery XI, 2005, 5806:298-306
[3]王晓飞, 张钧萍, 张晔.高光谱图像混合像元分解算法[J].红外与毫米波学报, 2010, 29(3):210-215
[4]Boardman J W, Kruse F A, Green R O.Mapping target signatures via partial unmixing of AVIRIS data in Summaries[J].Fifth JPL Airborne Earth Science Workshop, 1995, :23-26
[5]Winter M E.N-FINDR: an algorithm for fast autonomous spectral end-member determination in hyperspectral data[J].SPIE Proceedings of Imaging Spectrometry V, 1999, 3753:266-275
[6]Gruninger J H, Ratkowski A J, Hoke M L.The sequential maximum angle convex cone (SMACC) endmember model[J].SPIE Proceedings of Algorithms for Multispectral, Hyperspectral and Ultraspectral Imagery X, 2004, 5425:1-14
[7]Nascimento J M P, Dias J M B.Vertex component analysis: a fast algorithm to unmix hyperspectral data[J].IEEE Transactions on Geoscience and Remote Sensing, 2005, 43(4):898-910
[8]赵春晖，王楠楠.基于背景抑制及顶点成分分析的高光谱异常小目标检测[J].应用科技, 2009, 36(9):11-14
[9]Chang C I, Wu C C, Liu W M.A new growing method for simplex-based endmember extraction algorithm[J].IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(10):2804-2819
[10]张兵, 高连如.高光谱图像分类与目标探测[J].北京: 科学出版社, 2011, :102-150
[11]Candes E J.Compressive sampling[J].Proceedings of the International Congress of Mathematicians, 2006, 3:1433-1452
[12]Candes E J, Romberg J, Tao T.Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information[J].IEEE Trans. on Information Theory, 2006, 52(2):489-509
[13]Candes E J, Tao T.Near optimal signal recovery from random projections: Universal encoding strategies[J].IEEE Trans. Info. Theory, 2006, 52(12):5406-5425
[14]Donoho D L, Tsaig Y.Extensions of compressed sensing[J].Signal Processing, 2006, 86(3):533-548
[15]Donoho D L.Compressed sensing[J].IEEE Transactions on Information Theory, 2006, 52(4):1289-1306
[16]Candes E J.The restricted isometry property and its implications for compressed sensing[J].Comptes Rendus Mathematique, 2008, 346(9):589-592
[17]Aharon M, Elad M, Bruckstein A M.K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transactions on Signal Processing, 2006, 54(11):4311-4322
[18]Kohonen T.Self-organized Formation of Topologically Correct Feature Maps[J].Biological Cybernetics, 1982, 43(1):59-69
[19]Tropp J, Gilbert A.Signal recovery from random measurements via orthogonal matching pursuit[J].Transaction on Information Theory, 2007, 53(12):4655-4666
/
| 〈 |
|
〉 |