新技术新工艺
新技術新工藝
신기술신공예
New Technology & New Process
2015年
8期
61-62,63
,共3页
Gabor 小波%特征提取%年龄识别
Gabor 小波%特徵提取%年齡識彆
Gabor 소파%특정제취%년령식별
Gabor wavelets%feature extraction%age recognition
年龄识别是基于人脸的面部特征做出分析。由于人脸是塑性变形体,以及人的情感的不可捉摸,使得识别效率受到限制。通过使用 Gabor 小波提取人脸图像的特征,并对这些特征进行 PCA 降维。降维后的特征进入智能单粒子神经网络分类器(Intelligence Single Particle Optimization Neural Network Classifier,ISPONNC ),对输入的人脸图像的年龄进行识别和分类。年龄识别分为3类(Baby、Young 和 Old),识别率约为85%。通过对面部年龄进行识别,希望能对建立人性化的人机交互系统有所贡献。
年齡識彆是基于人臉的麵部特徵做齣分析。由于人臉是塑性變形體,以及人的情感的不可捉摸,使得識彆效率受到限製。通過使用 Gabor 小波提取人臉圖像的特徵,併對這些特徵進行 PCA 降維。降維後的特徵進入智能單粒子神經網絡分類器(Intelligence Single Particle Optimization Neural Network Classifier,ISPONNC ),對輸入的人臉圖像的年齡進行識彆和分類。年齡識彆分為3類(Baby、Young 和 Old),識彆率約為85%。通過對麵部年齡進行識彆,希望能對建立人性化的人機交互繫統有所貢獻。
년령식별시기우인검적면부특정주출분석。유우인검시소성변형체,이급인적정감적불가착모,사득식별효솔수도한제。통과사용 Gabor 소파제취인검도상적특정,병대저사특정진행 PCA 강유。강유후적특정진입지능단입자신경망락분류기(Intelligence Single Particle Optimization Neural Network Classifier,ISPONNC ),대수입적인검도상적년령진행식별화분류。년령식별분위3류(Baby、Young 화 Old),식별솔약위85%。통과대면부년령진행식별,희망능대건립인성화적인궤교호계통유소공헌。
Age recognition means making the analysis based on facial features.Since face is the plastic deformation and the unpredictable of human emotion,the recognition efficiency is limited.Use Gabor wavelet to extract the feature of the face image and to reduce the dimension of the characteristics.After the dimensionality reduction,the characteristics are put into ISPONNC (Intelligence Single Particle Optimization Neural Network Classifier)to recognize and classify the age of the face image.Age recognition is divided into three categories (Baby,Young and Old),and the recognition rate is about 85%. Through the way of experiment on face age recognition,it is done to contribute to build humanized human-computer interac-tion system.
