| 基于Gabor脸和隐马尔可夫模型的人像识别算法 |
| |
| 引用本文： | 曹林,王东峰,刘小军,邹谋炎.基于Gabor脸和隐马尔可夫模型的人像识别算法[J].高技术通讯,2005,15(6):25-29. |
| |
| 作者姓名： | 曹林 王东峰 刘小军 邹谋炎 |
| |
| 作者单位： | 1. 中国科学院电子学研究所,北京,100080;北京信息工程学院信息与通信工程系,北京,100101
2. 中国科学院电子学研究所,北京,100080 |
| |
| 基金项目： | 国家自然科学基金(60072020)，中国科学院科技创新基金(102107)资助项目 |
| |
| 摘 要： | 提出了基于Gabor小波变换和隐马尔可夫模型的人像识别算法。该算法先对人脸图像进行多分辨率的Gabor小波变换，采用主元分析法对每个结点进行降维，最后形成Gabor脸。把Gabor脸的每个特征结作为观测向量，对隐马尔可夫模型进行了训练，并把优化的模型参数用于人脸识别。实验结果表明，本文方法识别率高，复杂度较低，对部分遮挡的图像具有较大的容忍度。
|
| 关 键 词： | 隐马尔可夫模型 识别算法 Gabor小波变换 人像 主元分析法 多分辨率 人脸图像 观测向量 人脸识别 模型参数 识别率 复杂度 降维 结点 遮挡 |
Face recognition based on Gaborfaces and hidden Markov model |
| |
| Cao Lin,Wang Dongfeng,Liu Xiaojun,Zou Mouyan.Face recognition based on Gaborfaces and hidden Markov model[J].High Technology Letters,2005,15(6):25-29. |
| |
| Authors: | Cao Lin Wang Dongfeng Liu Xiaojun Zou Mouyan |
| |
| Abstract: | A new face recognition algorithm based on Gabor wavelets transform and hidden Markov model(HMM) is proposed. A bank of well-chosen Gabor filters are applied on the face images to construct a group of vectors called nodes, and then feature nodes are derived by using principal component analysis, which decrease the dimension of each node. The image including feature nodes is called Gaborfaces. A set of images representing different instances of the same person are used to train each HMM, and each individual in the database is represented by an optimal HMM face model. Experimental results show that the proposed algorithm has a high recognition rate with relatively low complexity and can keep a good tolerance to the partially occluded face images. |
| |
| Keywords: | face recognition Gaborfaces principal component analysis hidden Markov model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
