基于迭代光谱混合分析的高光谱图像解混技术分析-analysis of hyperspectral image unmixing technology based on iterative spectral mixing analysis.docxVIP

AbstractForthelowspatial resolution, mixed pixels areprevalent in theHyperspectralImage（HSI），which seriouslyaffects the featureidentificationand classification accuracy. To solvetheproblem, Spectral Unmixingproposed to identifyfeaturetypes and estimate abundance fraction is aquantitative remotesensingtechnology.The raditional spectral unmixingmethod devided into endmember extraction for and pixel unmixingis based on accuracyendmembers that construct mixed pixel, on the contrary, blind unmixingdirectlyaccess endmember and abundance without theinformation.This paper analyzesandcompares thepros and cons oftraditional spectralunmixingand blind unmixing, proposesaHyperspectral imageunmixingmethod baseoniterativespectral mixture analysis in combination ofthetowmethods.At first, a common spectral mixturemodel and apixel unmixingmethod with excessive endmembers areintroduced in this paper, then themixed pixel decomposition method based on iterativespectral mixture analysis is validatedwith real hyperspectral data;Second, non-negative matrixfactorization is analyzed, andweighted non-negativematrixfactorization whose weighted matrixis settled is proposed as aBlind Unmixingalgorithm; Finally, thenon-negativematrix factorization algorithm conbines themixed pixeldecomposition method based on iterative spectral mixture analysis algorithm, the endmember extraction algorithm called iterativeerror analysis,and the endmemberupdatemethod based on statistical endmemberto ablind unmixing method which is validated with real and simulation hyperspectral data.Keywords:Hyperspectral, Pixel unmixing, Non-negativeMatrixFactorization,Iterative Spectral MixtureAnalysis目录摘要...........................................................................................................IABSTRACT...........................................................................................II1 绪论1.1研究背景及意义................................................................................11.2光谱解混的现状及其发展趋势.......