| 二进制粒子群算法在人脸识别中的应用 |
| |
| 引用本文： | 程国建,石彩云,朱凯.二进制粒子群算法在人脸识别中的应用[J].计算机工程与设计,2012,33(4):1558-1562. |
| |
| 作者姓名： | 程国建 石彩云 朱凯 |
| |
| 作者单位： | 1. 西安石油大学计算机学院,陕西西安,710065
2. 美国太平洋大学工程与计算机科学系,美国加州斯托克顿CA 95211 |
| |
| 基金项目： | 国家自然科学基金项目(40872087) |
| |
| 摘 要： | 把二进制粒子群优化算法(BPSO)应用到人脸识别中.对人脸图像进行二维离散余弦变换(DCT),获得人脸图像的特征向量,应用BPSO算法对得到的特征向量进行特征选择,得到最具代表性的人脸特征.与遗传算法(GA)相比,在选择的特征较少的情况下,BPSO算法比遗传算法有更好的识别率.实验结果表明,BPSO算法应用到人脸识别中有较高的识别率,是一种非常有效的特征提取方法.
|
| 关 键 词： | 人脸识别 二进制粒子群优化算法 离散余弦变换 遗传算法 特征提取 |
Binary particle swarm optimization algorithm for human face recognition |
| |
| CHENG Guo-jian
,
SHI Cai-yun
,
ZHU Kai.Binary particle swarm optimization algorithm for human face recognition[J].Computer Engineering and Design,2012,33(4):1558-1562. |
| |
| Authors: | CHENG Guo-jian SHI Cai-yun ZHU Kai |
| |
| Affiliation: | 1.School of Computer Science,Xi’an Shiyou University,Xi’an 710065,China; 2.School of Engineering and Computer Science,University of the Pacific,Stockton,CA 95211,USA) |
| |
| Abstract: | The binary particle swarm optimization(BPSO) algorithm is introduced for face recognition.Firstly,the original face images are changed into feature vectors by two-dimensional discrete cosine transform(DCT).Secondly,the features are selected by the BPSO algorithm from the feature vectors,in order to get the most representative features of human faces.Compared with genetic algorithms(GA),the BPSO algorithm can obtain a higher recognition rate by means of a few of feature.The results show that the BSPO algorithm has a high recognition rate for human face recognition applications.The BSPO algorithm is certificated as an effective feature selection method. |
| |
| Keywords: | human face recognition binary particle swarm optimization algorithms discrete cosine transform genetic algorithms feature selection |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
