| 人耳人脸特征融合在身份鉴别中的研究 |
| |
| 引用本文： | 穆志纯,敦文杰.人耳人脸特征融合在身份鉴别中的研究[J].计算机科学,2009,36(5):251-253. |
| |
| 作者姓名： | 穆志纯 敦文杰 |
| |
| 作者单位： | 北京科技大学信息工程学院,北京,100083 |
| |
| 基金项目： | 国家自然科学基金，北京市教育委员会重点学科共建项目 |
| |
| 摘 要： | 针对单一人耳识别对姿态变化鲁棒性较差的问题,鉴于人脸在图像性质和生理位置上与人耳具有相似性和互补性,使用了多模态特征融合的方法提高姿态变化下的识别率.与传统的独立成分分析首先获得独立的基向量(ICAl)不同,提出了利用ICA直接获得独立的鉴剐特征的方法(ICA2).在USTB图像库上分别将两种ICA特征进行单模态和多模态的融合.实验表明,两种特征的融合提高了单一模态的识别率,并且多模态识别优于单一的人耳或人脸识别.
|
| 关 键 词： | 人耳识别 人脸识别 独立成分分析 特征融合 |
| 收稿时间： | 6/4/2008 12:00:00 AM |
Human Recognition Based on Feature Fusion of Ear and Face |
| |
| MU Zhi-chun DUN Wen-jie.Human Recognition Based on Feature Fusion of Ear and Face[J].Computer Science,2009,36(5):251-253. |
| |
| Authors: | MU Zhi-chun DUN Wen-jie |
| |
| Affiliation: | School of Information Engineering;University of Science and Technology Beijing;Beijing 100083;China |
| |
| Abstract: | The performance of single ear recognition is very poor when there is a pose variation.Because the properties of ear and face images are similar and the positions of the two features are complimentary,we improve the recognition rate using multi-modal feature fusion method.Aiming at difference from conventional independent Component Analysis(ICA1),we proposed a deformation of it to obtain statistical independent feature vectors directly(ICA2).The results of ICA feature fusion on the USTB database show that th... |
| |
| Keywords: | Ear recognition Face recognition Independent component analysis Feature fusion |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
点击此处可从《计算机科学》下载全文 |
