实用人脸识别系统的本征脸法实现(英文)  

PRACTICAL FACE RECOGNITION SYSTEM USING EIGENFACE ALGORITHM

作　　者：陈刚[1] 戚飞虎[1] 

机构地区：[1]上海交通大学计算机科学与工程系,上海200030

出　　处：《红外与毫米波学报》2000年第6期401-406,共6页Journal of Infrared and Millimeter Waves

基　　金：国防预研基金!(96 .J2 .4.2 )资助项目&&

摘　　要：将人脸模式的多个样本作为子模式 ,并将较多的人脸模式部分相交地分组 ,以减小计算量和便于识别系统的扩展 .在结合方法上采用基于贝叶斯理论的和结合规则 .使用 ORL 和 AR人脸图像库的人脸识别实验表明 ,本征脸法在采用多样本训练后 ,获得了较高的识别率和较短的识别时间 ;分组训练使识别系统可采用分布并行计算 ,从而减少训练时间 。A method taking multi samples as sub modes and grouping face modes into partial intersection ones was proposed to reduce computation and improve system extension property. In combination, the sum rule based on Bayesian theory was used. The face recognition experiments with the ORL and AR face databases showed that the eigenface algorithm using multi samples reached a high recognition rate and a reasonable time cost. Grouping face modes makes training become a distributed computation job which reduces time cost for training and brings the convenience of system extension when new face modes are to be added.

关 键 词：本征脸法 人脸识别系统 贝叶斯理论 识别时间 识别率 分组训练 人脸图像库 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]