步態(tài)識別論文基于人體骨架提取的步態(tài)分析

步態(tài)識別論文：基于人體骨架提取的步態(tài)分析步態(tài)識別論文:基于人體骨架提取的步態(tài)分析【中文摘要】近年來,隨著社會對敏感場合安全需求的提高,生物特征識別作為一種身份鑒別技術(shù),在全球范圍得到了大力的研究和發(fā)展。相比于傳統(tǒng)的身份識別MATLAB7.0面:首先,采用背景減除法從圖像序列中提取出運(yùn)動目標(biāo)輪廓,并運(yùn)用形態(tài)學(xué)運(yùn)算對過對人體骨架下肢運(yùn)動的分析,提取步態(tài)周期內(nèi)的下肢角度值變化作為步態(tài)特征。步態(tài)是具有非常大潛力的生物特征識別技術(shù),目前關(guān)于步態(tài)識別的研究主要還是處發(fā)現(xiàn),本文基于人體骨架的步態(tài)分析,能提取出人體的步態(tài)特征,對于個體識別系統(tǒng)的完善具有相當(dāng)大的意義。通過實(shí)驗(yàn),進(jìn)一步說明了步態(tài)識別技術(shù)的研究價值和光明前景?！居⑽恼縒ithagrowingneedforsecurityinsecurity-sensitiveenvironment,biometricsrecognition,asahumanidentificationapproach,hasbeengreatlyresearchedanddeveloped.ComparedtothetraditionalIDrecognitionmethod,thesecond-generationbiometricrecognitiontechnologybasedonvisionmovementcanidentifyindividualsbytheiruniquewalkingmanners.Gaithasitsownadvantageswhicharelong-distant,untouched,nonbody-invadinganddifficulttoconceal,isveryattractiveinthefieldofvisualsurveillance.BasedonagaitvideoclipandapplyingMATLAB7.0softwaretoprocessingimages,thegaitrecognitionmethodofusingthehumanskeletonisproposedinthispaper,whichmainlycoversthefollowingissues:Firstly,thebackgroundsubtractionisusedingaitdetectiontorecoverthemovingobjectafterscanningtheoriginalimage.Principalcurvescanreflecttheinherentstructureofthedataanddescribenonlineardata,whichisbeneficialtothecontourextraction.Secondly,Gaitcycleisanalyzed,thenwidthandheightofbodyanalysisisperformedtocomputerit.Lastly,basedontheideathatjoint-angletrajectoriesofbodypartsinwalkingmotionincludesufficientdynamicidentityinformation,agaitrecognitionmethodbasedonlower-limbmotionanalysisanddynamictimewrappingisproposed.Foreachgaitsequence,accordingtotheknowledgeofhumanbodyanatomy,thecoordinatesoflower-limbjointsareobtainedbyanalyzinglower-limbmotion,andthenthetrajectoriesoflower-1imbanglesinonecycleareextractedasfeaturevectors.Undoubtedly,gaitisaquitepotentialbiometric.However,relevantstudymainlystaysatinthephaseoftheoryresearchatpresent,andinordertoapplythetechnologyintopractice,therearestillalotofurgentproblemsneededtoberesolved.Experimentalresultsthatthecharacteristicofthehumangaitcanbeidentifiedbasedontheanalysisofhumanskeleton,whichhassignificantimportancefortheimprovementofindividualrecognition.Experimentsfurthershowtheresearchvalueandthebrightfutureofthegaitrecognition.【關(guān)鍵詞】步態(tài)識別細(xì)化人體骨架關(guān)節(jié)角度【英文關(guān)鍵詞】gaitrecognitionthinhumanskeletonjointangle【備注】索購全文在線加好友:.3.8848【目錄】基于人體骨架提取的步態(tài)分析摘要5-6Abstract6第一章緒論9-151.1步態(tài)識別簡介9-101.2步態(tài)識別的理論依據(jù)和發(fā)展?fàn)顩r10-121.3步態(tài)識別系統(tǒng)12-131.4論文的組織結(jié)構(gòu)13-15第二章步態(tài)識別技術(shù)介紹15-282.1MATLAB7.0簡介15-182.1.1圖像和MATLAB15-162.1.2MATLAB的特色和應(yīng)用16-182.2分割運(yùn)動目標(biāo)18-192.2.1時域差分法182.2.2背景減除法18-192.2.3光流法192.3步態(tài)特征提取19-232.3.1基于模型的方法19-222.3.2基于統(tǒng)計的方法22-232.4分類23-272.4.1分類識別方法23-252.4.2分類數(shù)據(jù)庫25-272.5本章小結(jié)27-28第三章運(yùn)動目標(biāo)分割與形態(tài)學(xué)運(yùn)算28-383.1運(yùn)動目標(biāo)分割28-333.1.1運(yùn)動背景的建模29-313.1.2特定閾值下的圖像差分31-333.2形態(tài)學(xué)運(yùn)算處理33-363.2.1膨脹34-353.2.2腐蝕353.2.3基于膨脹和腐蝕的形態(tài)學(xué)處理及分析35-363.3輪廓線提取36-373.4本章小結(jié)37-38第四章步態(tài)周期內(nèi)骨架及下肢運(yùn)動變化分析38-584.1步態(tài)周期38-404.1.1步態(tài)周期定義38-394.1.2步態(tài)周期提取39-404.2細(xì)化與人體骨架的獲取40-434.2.140-424.2.2人體骨架的定義42-434.2.3人體骨架的獲取434.3下肢運(yùn)動分析及特征提取43-504.3.1下肢關(guān)節(jié)點(diǎn)角度分析44-464.3.2人體骨架下肢特征提取46-504.4下肢特征提取方