预览加载中,请您耐心等待几秒...
1/3
2/3
3/3

在线预览结束,喜欢就下载吧,查找使用更方便

如果您无法下载资料,请参考说明:

1、部分资料下载需要金币,请确保您的账户上有足够的金币

2、已购买过的文档,再次下载不重复扣费

3、资料包下载后请先用软件解压,在使用对应软件打开

基于线性最小二乘支持向量机的光谱端元选择算法 Abstract Spectralunmixingisafundamentalissueinhyperspectralimageanalysis,whichaimstoestimatetheabundancefractionsofdifferentpurematerials,alsoknownasendmembersorspectralsignatures,presentineachpixelofahyperspectralimage.Endmemberselectionisacrucialstepinspectralunmixingtoidentifyendmembersaccurately.Thispaperproposesanewendmemberselectionalgorithmbasedonlinearleast-squaressupportvectormachines(LSSVM).Theproposedalgorithmistestedonseveralsyntheticdatasetsandrealhyperspectralimages,andtheresultsshowthattheproposedalgorithmisaneffectivemethodforendmemberselection. Introduction Hyperspectralimagingtechnologyhasbeenwidelyusedinvariousfields,suchasagriculture,environment,remotesensing,andmedicine.Hyperspectralimagestypicallyconsistofhundredsofspectralbands,whichcorrespondtodifferentwavelengthsoflight.Inordertointerpretthespectralinformation,itisnecessarytoidentifythepurematerialsorendmembersthatarepresentineachpixeloftheimage.Spectralunmixingisacommontechniqueusedtoestimatetheabundancefractionsofdifferentendmembersineachpixeloftheimage. Endmemberselectionisacrucialstepinspectralunmixing,asithasasignificantimpactontheaccuracyofendmemberidentificationandabundanceestimation.Therefore,itisessentialtodevelopeffectiveandefficientalgorithmsforendmemberselection. Inrecentyears,manyalgorithmshavebeenproposedforendmemberselection,includingN-FINDR,PPI,VCA,andATGP.However,thesealgorithmshavesomelimitations,suchassensitivitytonoise,highcomputationalcomplexity,andtherequirementofpriorknowledgeaboutthenumberofendmembers.Therefore,itisnecessarytodevelopnewendmemberselectionalgorithmsthatcanovercometheselimitations. Inthispaper,weproposeanewendmemberselectionalgorithmbasedonlinearleast-squaressupportvectormachines(LSSVM).Theproposedalgorithmcaneffectivelyselecttheendmemberswhileovercomingthelimitationsofexistingalgorithms. RelatedWork Endmemberselectionisanessentialstepinhyperspectralimageanalysis,andmanyalgorithmshavebeenproposedforthistask.Inthissection,webrieflyr