基于金字塔分層模型的多光譜目標識別.pdf

1、聲明本學(xué)位論文是我在導(dǎo)師的指導(dǎo)下取得的研究成果，盡我所知，在本學(xué)位論文中，除了加以標注和致謝的部分外，不包含其他人已經(jīng)發(fā)表或公布過的研究成果，也不包含我為獲得任何教育機構(gòu)的學(xué)位或?qū)W歷而使用過的材料。與我一同工作的同事對本學(xué)位論文做出的貢獻均已在論文中作了明確的說明。研究生簽名：曼堡珈廠7年弓月弓日學(xué)位論文使用授權(quán)聲明南京理工大學(xué)有權(quán)保存本學(xué)位論文的電子和紙質(zhì)文檔，可以借閱或上網(wǎng)公布本學(xué)位論文的部分或全部內(nèi)容，可以向有關(guān)部門或機構(gòu)送交并授

2、權(quán)其保存、借閱或上網(wǎng)公布本學(xué)位論文的部分或全部內(nèi)容。對于保密論文，按保密的有關(guān)規(guī)定和程序處理。研究生簽名：墨翼加一年弓月弓日碩士學(xué)位論文AbstractWiththehighspeeddevelopmentofartificialintelligence，generaltargetdetectionforaclassofspecifictargetsplaysavitalpartinthemachinevisionThosetradit

3、ionalclassifiermodels，howeverrequirealotoftrainingsamples，whichmayleadtoslowlearningprocessWhat’Smore，phenomenaofoverfittrainingparametersmayeasilyOCCHLThereforenewwaysoftargetdetectionwithouttraininghavebeendevelopedinr

4、ecentyearsThispaperfocusesonthemultispectralimagedetection，andputsforwardtworobusttargetdetectionmodel(1)PyramidhierarchicalstructuremodelbasedontheLARKInthismodelweconstructthe3D—LARKfeaturedescriptorsbyfullyusingthespa

5、tialstructureandtheabundantspectralinformation，inordertosolvethelowrecognitionaccuracycausedbythelossofspectralinformationingrayimagesMeanwhile，forthecomplexcalculationandlowaccuracyweproposethepyramidhierarchicalmodelFo

6、rthechangeabletargetsinimages，acompositetemplatesetisestablishedtoensurealltargetsCanbedetected，whichismultiscaleandmultiangleExperimentsshowthat，comparedwimthetraining—freealgorithmforgrayimagestheproposedmodelusesthesp

7、aceandspectruminformationeffectivelyandreducestheinterferenceofthelocalsimilarstructureinthebackgroundnoise，aswellastheerrordetectionrate(2)TargetdetectionmodelbasedonthelocalstructureandneighborconstraintReliedontheabov

8、emodel，theneighborstructurecharacteristicisfurtherintroducedtooptimizetheintegrityofthedetectionresultsWhatsmore，淅ththeneighborstructurecharacteristicand3DLARKfeaturedescriptors，thealternatingdetectionsisconductedinthepy

9、ramidmodeltoimprovethedetectionaccuracNSOastheefficiencyExperimentsaretakeninthestandardlibrarywithCarsandpedestrian，andcompared誦thothertrainingfreealgorithmItcanbeprovedthat，theproposedmodeliseffectiveincolorimagetarget