基于小波變換及極大模方法的面容特征抽取--一種適用于面容知識(shí)的雙正交小波的探討.pdf

1、山東大學(xué)碩士學(xué)位論文基于小波變換及極大模方法的面容特征抽取一種適用于面容知識(shí)的雙正交小波的探討姓名：段秉乾申請(qǐng)學(xué)位級(jí)別：碩士專業(yè)：運(yùn)籌學(xué)與控制論指導(dǎo)教師：關(guān)華勇2000.5.1Abstract：HumanfacerecognitiontechnologyisoneofthenecessarycomponentsincreaturerecognitiontechnologyTheresearchonithasgreatvalueincom

2、puterscienceandcanbeappliedonmanyapplicationsTheextractionoffacialfeaturewilleffecttheaccuracyofthehumanfacerecognitiondirectlySOitplaysanimportantroleinthehumanfacerecognitionInthispaperproposeabaseofbiorthogonalwavelet

3、sforwaveletdecompositionofhumanfaceimage；improvethewaveletmaximamoduleedgedetectionalgorithmandthenewalgorithmiscalledwaveletmaximamoduleareaedgedetectionalgorithm；proposeahumanfacialfeatureextractionalgorithmThefeaturee

4、xtractionalgorithmiscomposedofthreemainstage：1)waveletdecompositionofatargetimage2)backgroundsegmentationandfaceregionlocation3)facialfeatureextractionInthewaveletdecompositionofatargetimagestage，themallat’Sdecomposition

5、algorithmandtheproposedbaseofbiorthogonalwaveletsareadoptedInthebackgroundsegmentationandfaceregionlocationstage，thewaveletsmaximamoduleareaedgedetectionalgorithmisadoptedtosegmentthepanetexturebackgroundAndfaceregionoft

6、hetargctimageisestimatedIIIthefeatureextractionstagethewaveletmaximamoduleareaedgedetectionalgorithmisadoptedtodetectedgesinfaceregionAndedgestructureanalysismethodisadoptedtoextractfacialfeatureItisshownbythesimulationr

7、esultsthattheproposedalgorithmsCanautomaticallyandexactlyextractthefacialfeatureKeywords：facialfeatureextraction，biothogonalwavelet，wavelettransformlinephase，maximamodule，mallat’salgorithm，edgedetection，panetexturebackgr