正在加载图片...
第8卷第6期 智能系统学报 Vol.8 No.6 2013年12月 CAAI Transactions on Intelligent Systems Dec.2013 D0:10.3969/j.issn.1673-4785.201301005 网络出版地址:http://www.cnki.net/kcms/detail/23.1538.TP.20131012.1814.003.html 基于特征融合的人脸图像性别识别 李同宇,李卫军,覃鸿 (中国科学院半导体研究所高速电路与人工神经网络实验室,北京100083)》 摘要:在人脸图像的性别识别方法研究中,存在同一个人既参与训练又参与测试的情况,所得结论有一定的局限 性针对此问题,建立相互独立的测试集和训练集传统性别识别模型,受相关参数影响较大,稳定性有待提高,为此, 提出一种基于特征融合的人脸图像性别识别方法,采用主成分分析和正交化的线性判别分析相结合的方法表述图 像的全局特征,突破传统线性判别分析二分类时秩的限制,采用均衡的局部二值模式方法表述图像的局部特征,将 少量全局特征和局部特征相融合,形成人脸图像的性别特征支持向量机用于实现性别特征的分类实验结果表明, 此方法在具有一定稳定性的同时,能获得较高的识别率。 关键词:人脸识别:性别识别:全局特征;局部特征;特征融合 中图分类号:TP181文献标志码:A文章编号:1673-4785(2013)06-0505-07 中文引用格式:李同字,李卫军,覃鸿.基于特征融合的人脸图像性别识别[J].智能系统学报,2013,8(6):505-511. 英文引用格式:LI Tongyu,LI Weijun,QIN Hong.Facial image gender recognition method based on feature fusion[J].CAAI Transactions on Intelligent Systems,2013,8(6):505-511. Facial image gender recognition method based on feature fusion LI Tongyu,LI Weijun,QIN Hong (Laboratory of High-speed Circuit Artificial Neural Network,Institute of Semiconductors,CAS,Beijing 100083,China) Abstract:In the research on the facial images gender recognition method,there exist cases where a person attends both training and testing,therefore,the conclusions attained are restrictive.In order to solve this problem,a mutu- ally independent testing set and training set have been established;the traditional gender recognition model is great- ly affected by the relevant parameters and the stability needs to be increased,therefore,a novel gender recognition method utilizing facial images and based on feature fusion has been proposed.With this method,the analysis of the main components and the orthogonal linear distinguishing analysis are combined to describe the overall features of the facial image,the restriction of order in binary classification for traditional linear distinguishing analysis is bro- ken through,a balanced local binary value pattern method is applied to describe the local features of the image, and some of the overall features and the local features are fused to form the gender features of the facial image.The support vector machine is used for the classification of gender characteristics.The experimental results demonstrate that the proposed method ensures both a high recognition rate and robustness. Keywords:face recongnition;gender recongnition;overall feature;local feature;feature fusion 人脸是人类生物特征的重要组成部分,人脸图 技术的飞速发展,逐渐形成了人脸识别领域的经典 像分类是人脸研究的主要方向,基于人脸图像的性 人脸图像库,如FERET人脸库、ORL人脸库和CAS 别识别是人脸图像分类的典型应用.随着人脸识别 PEAL人脸库,这些人脸库都包含个人的多姿态、多 光照的图像,属于多样本人脸数据库山.越来越多的 收稿日期:2013-01-06.网络出版日期:2013-10-12 基金项目:国家自然科学基金重大研究计划资助项目(90920013). 关于性别识别的研究直接使用经典的人脸图像库, 通信作者:李卫军.E-mail:wjli@semi.ac.cm. 同一个人既参与训练,又参与测试,识别率不断提第 8 卷第 6 期 智 能 系 统 学 报 Vol.8 №.6 2013 年 12 月 CAAI Transactions on Intelligent Systems Dec. 2013 DOI:10.3969 / j.issn.1673⁃4785.201301005 网络出版地址:http: / / www.cnki.net / kcms/ detail / 23.1538.TP.20131012.1814.003.html 基于特征融合的人脸图像性别识别 李同宇,李卫军,覃鸿 (中国科学院半导体研究所 高速电路与人工神经网络实验室,北京 100083) 摘 要:在人脸图像的性别识别方法研究中,存在同一个人既参与训练又参与测试的情况,所得结论有一定的局限 性.针对此问题,建立相互独立的测试集和训练集.传统性别识别模型,受相关参数影响较大,稳定性有待提高,为此, 提出一种基于特征融合的人脸图像性别识别方法,采用主成分分析和正交化的线性判别分析相结合的方法表述图 像的全局特征,突破传统线性判别分析二分类时秩的限制,采用均衡的局部二值模式方法表述图像的局部特征,将 少量全局特征和局部特征相融合,形成人脸图像的性别特征.支持向量机用于实现性别特征的分类.实验结果表明, 此方法在具有一定稳定性的同时,能获得较高的识别率. 关键词:人脸识别;性别识别;全局特征;局部特征;特征融合 中图分类号: TP181 文献标志码:A 文章编号:1673⁃4785(2013)06⁃0505⁃07 中文引用格式:李同宇,李卫军,覃鸿. 基于特征融合的人脸图像性别识别[J]. 智能系统学报, 2013, 8(6): 505⁃511. 英文引用格式:LI Tongyu, LI Weijun, QIN Hong. Facial image gender recognition method based on feature fusion[ J]. CAAI Transactions on Intelligent Systems, 2013, 8(6): 505⁃511. Facial image gender recognition method based on feature fusion LI Tongyu, LI Weijun, QIN Hong (Laboratory of High⁃speed Circuit & Artificial Neural Network, Institute of Semiconductors, CAS, Beijing 100083, China) Abstract:In the research on the facial images gender recognition method, there exist cases where a person attends both training and testing, therefore, the conclusions attained are restrictive. In order to solve this problem, a mutu⁃ ally independent testing set and training set have been established; the traditional gender recognition model is great⁃ ly affected by the relevant parameters and the stability needs to be increased, therefore, a novel gender recognition method utilizing facial images and based on feature fusion has been proposed. With this method, the analysis of the main components and the orthogonal linear distinguishing analysis are combined to describe the overall features of the facial image, the restriction of order in binary classification for traditional linear distinguishing analysis is bro⁃ ken through, a balanced local binary value pattern method is applied to describe the local features of the image, and some of the overall features and the local features are fused to form the gender features of the facial image. The support vector machine is used for the classification of gender characteristics. The experimental results demonstrate that the proposed method ensures both a high recognition rate and robustness. Keywords:face recongnition; gender recongnition; overall feature; local feature; feature fusion 收稿日期:2013⁃01⁃06. 网络出版日期:2013⁃10⁃12. 基金项目:国家自然科学基金重大研究计划资助项目(90920013). 通信作者:李卫军. E⁃mail:wjli@ semi.ac.cn. 人脸是人类生物特征的重要组成部分,人脸图 像分类是人脸研究的主要方向,基于人脸图像的性 别识别是人脸图像分类的典型应用.随着人脸识别 技术的飞速发展,逐渐形成了人脸识别领域的经典 人脸图像库,如 FERET 人脸库、ORL 人脸库和 CAS⁃ PEAL 人脸库,这些人脸库都包含个人的多姿态、多 光照的图像,属于多样本人脸数据库[1] .越来越多的 关于性别识别的研究直接使用经典的人脸图像库, 同一个人既参与训练,又参与测试,识别率不断提
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有