投影尋蹤指標(biāo)小波估計(jì)與CDA降維方法.pdf

1、西北工業(yè)大學(xué)碩士學(xué)位論文投影尋蹤指標(biāo)小波估計(jì)與CDA降維方法姓名：鄭鵬申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：應(yīng)用數(shù)學(xué)指導(dǎo)教師：田錚20060301AbstractProjectionpursuitinstatisticsandsignalprocessinghasreceivedattentionbecauseofitspotentialapplicationsThepurposeofprojectionpursuitistodiscoverinte

2、restinglowdimensionalprqiectionsfromhighdimensionaldatasetbynumericallymaximizingaprojectionindexTherefore，thechoiceofprojectionindexisveryimportantinprojectionpursuitCurvilinearDistanceAnalysis(CDA)isanonlinearprojectio

3、nmethodthatCanpreservethetopologyofinitialdataInrecentyears，manyauthorshaveshownagooddealofinterestinilThispaperproposedthewaveletkernelestimationforprojectionindicesanda|ilizedCDAtoreducethehyperspectraldataAndweobtaine

4、dseveralresuItsasfollows：Inchapter2，weproposedthewaveletkernelestimationfortwoprojectionindices，andprovetheestimationsareasymptoticunbiasedestimationandconvergenceinthemeansquareInchapter3，wegiveSOlnestatisticalpropertie

5、softhemultivariatewaveletdcnshyestimate—orAndthenprotmsetheMultivariatewaveletkernelestimationforprojectionindexAtlastweobtainsomestatisticalpropertiestotheestimationInthelastchapterweobtainedanewalgoritluntoapply011theH

6、yperspectralimageunsupervisedclassificationTheCDAisappliedasapre—processingstepteducethedimensionallyofthedatasetAndthenICAMMisusedforunsupervisedclassificationKeywords：nonlinearprojection，projectionpursuitprojectioninde