本文選題：非負(fù)矩陣分解　切入點(diǎn)：高光譜圖像解混　出處：《西安電子科技大學(xué)》2014年碩士論文　論文類型：學(xué)位論文

【摘要】：高光譜遙感成像是一種新型的對(duì)地觀測(cè)技術(shù)。成像光譜儀能夠記錄數(shù)十至數(shù)百個(gè)波段的光譜信息，使得對(duì)各種類型地物的準(zhǔn)確分類與識(shí)別成為可能。然而，因?yàn)槎S平面的分辨率的約束作用，高光譜圖像中的一個(gè)像元可能含有幾種不同類型的地物，形成了所謂的“混合像元”，直接影響了后續(xù)地物的檢測(cè)與識(shí)別。光譜解混技術(shù)能夠?qū)ⅰ盎旌舷裨狈纸鉃閹追N基本類型的地物光譜向量(即端元)與其對(duì)應(yīng)的混合比例(即豐度)的乘積，從而獲得亞像元信息，進(jìn)而提高后續(xù)數(shù)據(jù)應(yīng)用的效果，最近在遙感領(lǐng)域得到了普遍關(guān)注。 非負(fù)矩陣分解（Nonnegative Matrix Factorization, NMF）將一個(gè)非負(fù)矩陣分解為兩個(gè)非負(fù)矩陣的乘積，其分解模型與混合像元解混模型相似，非常適于解決光譜解混問(wèn)題。然而，基于基本NMF算法的高光譜圖像解混算法是一個(gè)欠定問(wèn)題。為了得到唯一確定的解混結(jié)果，需要加入正則約束消除其欠定性，針對(duì)該問(wèn)題，本論文在研究流形學(xué)習(xí)的基礎(chǔ)上，挖掘高光譜數(shù)據(jù)的先驗(yàn)信息，設(shè)計(jì)了多種流形正則下的非負(fù)矩陣分解解混技術(shù)。具體工作如下： (1)針對(duì)現(xiàn)有NMF解混算法僅利用了高光譜數(shù)據(jù)的光譜信息，忽略了高光譜數(shù)據(jù)空間信息的缺陷，設(shè)計(jì)了一種基于空-譜流形正則的NMF高光譜數(shù)據(jù)解混方法。構(gòu)造局部窗挖掘高光譜數(shù)據(jù)的空間信息，設(shè)計(jì)基于空譜流形正則的半監(jiān)督NMF解混算法。在人工合成高光譜數(shù)據(jù)和實(shí)際高光譜數(shù)據(jù)上進(jìn)行試驗(yàn)，對(duì)于S3NMF的各個(gè)正則參數(shù)的選擇、算法的收斂性、對(duì)于解混出的端元和豐度進(jìn)行對(duì)比分析、算法的魯棒性分析和在含有不同端元數(shù)的高光譜數(shù)據(jù)解混結(jié)果分析結(jié)果表明：本方法在數(shù)值指標(biāo)方面、視覺(jué)效果方面和算法的魯棒性均優(yōu)于CNMF、GLNMF等方法。 (2)針對(duì)高光譜圖像特性中的稀疏先驗(yàn)，設(shè)計(jì)了一種基于稀疏多流形正則的非負(fù)矩陣分解高光譜數(shù)據(jù)解混方法。高光譜圖像中的混合像元僅由有限多種地物的光譜向量混合產(chǎn)生，基于這種稀疏性，以及稀疏編碼系數(shù)相近的像元具有相近豐度向量的假設(shè)，構(gòu)造稀疏流形正則，設(shè)計(jì)基于稀疏多流形正則的非負(fù)矩陣分解高光譜數(shù)據(jù)解混算法（MMSNMF）。實(shí)驗(yàn)中將與S3NMF等方法的高光譜解混結(jié)果進(jìn)行對(duì)比，在人工合成高光譜數(shù)據(jù)和實(shí)際高光譜數(shù)據(jù)上進(jìn)行試驗(yàn)，結(jié)果表明：本方法相比前面的方法解混的在數(shù)值和視覺(jué)方面的效果有了一定的提高。 (3)針對(duì)現(xiàn)有NMF解混算法未充分挖掘圖像空間信息的局限，設(shè)計(jì)了一種基于相似性流形正則的非負(fù)矩陣分解高光譜數(shù)據(jù)線性解混方法。采用鄰接權(quán)值計(jì)算權(quán)值的方法比用在K-近鄰中用熱核函數(shù)獲得更加準(zhǔn)確的像元空間關(guān)系信息，針對(duì)高光譜圖像中光譜之間相似性的特性，引入相似度函數(shù)對(duì)光譜信息的相似度特性進(jìn)行描述。構(gòu)造相似性正則，設(shè)計(jì)基于相似性流形正則的非負(fù)矩陣分解高光譜數(shù)據(jù)線性解混方法（SMNMF）。實(shí)驗(yàn)中將本方法與MMSNMF等方法的高光譜解混結(jié)果進(jìn)行對(duì)比，在人工合成高光譜數(shù)據(jù)和實(shí)際高光譜數(shù)據(jù)上進(jìn)行試驗(yàn)，結(jié)果表明：本方法相比前面方法，在解混的數(shù)值和視覺(jué)結(jié)果方面均有了一定的提高。 本文的工作得到了國(guó)家重點(diǎn)基礎(chǔ)研究發(fā)展計(jì)劃（973計(jì)劃）： No.2013CB329402，國(guó)家自然科學(xué)基金61072108，60601029，60971112，61173090)，，新世紀(jì)優(yōu)秀人才項(xiàng)目：NCET-10-0668，高等學(xué)校學(xué)科創(chuàng)新引智計(jì)劃（111計(jì)劃）：No. B0704，教育部博士點(diǎn)基金(20120203110005)，武器裝備預(yù)研基金項(xiàng)目(9*****)，以及華為創(chuàng)新研究計(jì)劃項(xiàng)目(IRP-2013-01-09)的資助。
[Abstract]:Hyperspectral imaging is a new technology of earth observation. The spectral information of imaging spectrometer to record tens to hundreds of bands, making the accurate classification and identification of various types of objects possible. However, because of the confinement effect of two-dimensional resolution, high optical spectrum of a pixel in the image may contain several different the type of the object, the formation of the so-called "mixed pixel", directly affects the detection and recognition of objects. The spectral unmixing technique can be "mixed pixel" is divided into several basic types of spectral vector (endmembers) corresponding to the mixing ratio (i.e., abundance) product, so as to obtain sub-pixel information then, to improve the application effect of follow-up data, recently in the field of remote sensing has been widely concerned.
Non negative matrix factorization (Nonnegative Matrix, Factorization, NMF) be a non negative matrix factorization is the product of two non negative matrices, the decomposition model and unmixing model similarity, is very suitable for solving spectral unmixing problem. However, based on the basic NMF algorithm for hyperspectral image unmixing algorithm is an underdetermined in order to get mixed results. The only solution to determine the need to add regular constraints to eliminate the underdetermined, aiming at this problem, this thesis research on manifold learning in hyperspectral data mining, prior information, design a variety of manifold is the non negative matrix factorization unmixing technology. The specific work is as follows:
(1) the existing NMF unmixing algorithm using only the spectral information of hyperspectral data, ignoring the defects of high spectral data of spatial information, design a solution of mixed NMF hyperspectral data in space and spectrum. Based on Manifold Regularization to construct a local window mining spatial information of hyperspectral data, the design of air manifold regularization spectrum semi supervised NMF algorithm based on mixed solution. Experiments were carried out in synthetic hyperspectral data and real hyperspectral data, for each of the regularization parameter selection of S3NMF, the convergence of the algorithm, the endmember and abundance of the mixed solution were analyzed, the robustness of the algorithm and Analysis on hyperspectral data with different end the number of element unmixing results analysis results show that this method in the numerical index, visual effects and robustness of the algorithm are better than those of CNMF, GLNMF and other methods.
(2) according to the sparsity characteristics of hyperspectral image in the design of a non negative matrix sparse Manifold Regularization Based on decomposition of high spectral data unmixing method. Mixed pixels in hyperspectral images is only produced by the spectral vector mixed finite variety of features, which based on sparse, and sparse pixel encoding similar coefficient with similar abundance vector hypothesis, construct sparse Manifold Regularization, design based on non negative matrix sparse Manifold Regularization decomposition of high spectral data unmixing algorithm (MMSNMF). The S3NMF with high spectral unmixing for comparison test in synthetic hyperspectral data and real hyperspectral data. The results show that the method of this method compared to previous mixing solutions in numerical and visual effect are improved.
(3) the existing NMF unmixing algorithm does not fully tap the image spatial information limitations, the design of a non negative matrix similarity canonical manifold decomposition based on linear unmixing method for hyperspectral data. The pixel spatial information method using adjacency weight calculation weights is more accurate than that used in the K- function was used in thermonuclear neighbor the in between spectral hyperspectral image similarity features, the similarity characteristics of the spectral information into the similarity function described. Structural similarity design based on regular, non negative matrix similarity canonical manifold decomposition of hyperspectral data linear unmixing method (SMNMF). In this experiment, high spectral method and MMSNMF method the unmixing results were compared in the experiment, synthetic hyperspectral data and real hyperspectral data. The results showed that this method compared to the previous method, the numerical and visual mixing solutions There is a certain improvement in the fruit.
This work was supported by the national basic research program (973 Program):No.2013CB329402, the National Natural Science Foundation 61072108606010296097111261173090), New Century Talents Project: NCET-10-0668, higher school subject innovation engineering plan (111 plan): No. B0704, Doctoral Fund of Ministry of Education (20120203110005), Armament Research Foundation (9*****). And HUAWEI innovation research project (IRP-2013-01-09) funding.

【學(xué)位授予單位】：西安電子科技大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：TP751

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 游曉斌,游先祥,相瑩瑩;混合像元及混合像元分析[J];北京林業(yè)大學(xué)學(xué)報(bào);2003年S1期

2 童慶禧;遙感科學(xué)技術(shù)進(jìn)展[J];地理學(xué)報(bào);1994年S1期

3 錢樂(lè)祥,泮學(xué)芹,趙芊;中國(guó)高光譜成像遙感應(yīng)用研究進(jìn)展[J];國(guó)土資源遙感;2004年02期

4 束炯;王強(qiáng);孫娟;;高光譜遙感的應(yīng)用研究[J];華東師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2006年04期

5 ;L_(1/2) regularization[J];Science China(Information Sciences);2010年06期

6 張斧;成像光譜技術(shù)發(fā)展的科學(xué)背景和現(xiàn)狀[J];激光與紅外;1994年01期

7 劉維湘;鄭南寧;游屈波;;非負(fù)矩陣分解及其在模式識(shí)別中的應(yīng)用[J];科學(xué)通報(bào);2006年03期

8 劉卓,易東云;投影尋蹤方法與高光譜遙感圖像數(shù)據(jù)特征提取的研究[J];數(shù)學(xué)理論與應(yīng)用;2003年01期

9 薛綺,匡綱要,李智勇;基于線性混合模型的高光譜圖像端元提取[J];遙感技術(shù)與應(yīng)用;2004年03期

10 李天宏,楊海宏,趙永平;成像光譜儀遙感現(xiàn)狀與展望[J];遙感技術(shù)與應(yīng)用;1997年02期

本文編號(hào)：1642164