| 基于形态学小波理论和SVM神经网络的人脸识别 |
| |
| 引用本文： | 李伟,彭玉峰.基于形态学小波理论和SVM神经网络的人脸识别[J].河南师范大学学报(自然科学版),2012,40(5):61-64. |
| |
| 作者姓名： | 李伟 彭玉峰 |
| |
| 作者单位： | 河南师范大学物理与信息工程学院,河南新乡,453007 |
| |
| 摘 要： | 主要研究了快速识别人脸的基本算法,它包括人脸检测和人脸识别两部分.人脸检测部分利用肤色电平的聚类特性和形态学处理检测出准人脸图像,再利用小波特征提取出特征进行人脸认证.人脸识别部分采用支持向量机(SVM)神经网络进行人脸识别.支持向量机神经网络对二类判别具有很强的识别能力.对于N类判别需连续使用N次.该方法识别速度快,且不受发型、头饰、眼镜等的影响.仿真证明了该方法的有效性.
|
| 关 键 词： | 形态学理论 小波变换 支持向量机神经网络 人脸识别 |
The Face Recognition Based on Morphology and Wavelet Theory with SVM Neural Network |
| |
| LI Wei
,
PENG Yu-feng.The Face Recognition Based on Morphology and Wavelet Theory with SVM Neural Network[J].Journal of Henan Normal University(Natural Science),2012,40(5):61-64. |
| |
| Authors: | LI Wei PENG Yu-feng |
| |
| Institution: | (College of Physics and Information Engineering,Henan Normal University,Xinxiang 453007,China) |
| |
| Abstract: | The basic algorithm is mainly investigated about fast face-recognition in the paper.It contains face detection and face recognition.In the mould about face detection,the quasi-face images can be detected using the skin cluster character and morphology theory,and then face features can be extracted using wavelet to get face authentication.In the mould about face recognition,SVM neural network has been adopted to recognize the faces.The SVM neural network shows good performance for two-class problems.N times can be used successively for N-class problems.The method owns fast speed for recognition and can not be affacted by hair-style,head ornaments and glasses etc. |
| |
| Keywords: | morphology wavelet transform SVM neural network face recognition |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
