摘要
针对人脸易受到年龄、表情等影响,提出了脸和耳相结合的组合识别方法。利用二维Fisher线性判别(2DFLD)方法分别进行了脸、耳图像层和特征层的组合识别。在北京科技大学人耳库和ORL人脸库上进行实验,结果表明,图像层组合和特征层组合的识别率分别为97.5%、95.0%,分别比人脸识别提高了12.5%和10.0%,比人耳识别提高了5.0%和2.5%;与同样应用于组合识别的主成分分析(PCA)、二维PCA(2DPCA)比较,也取得了较好识别效果。这说明,多生物特征组合识别是一种有效的识别方法。
Aiming at the limitation of facial biometrics due to the changes in features caused by age and facial expressions etc, a combination recognition method based on the face and ear is proposed. Two-dimensional fisher linear diseriminant（2DFLD） is utilized to carry on the recognition of facial and otie im- age fusion. By using the BUST ear databased and ORL face database, the experimental results indicate that the combination recognition rate of images fusion and features fusion reaches respectively 97.5 % and 95% ,which is 12.5% and 10% higher than the recognition rate of face,and 5% and 2.5% higher than the recognition rate of ear. The combination recognition based on the 2DFLD algorithm has a good recognition ability compared with the methods such as PCA, 2DPCA. This indicates that the multi-bio- logical features combination recognition is an effective recognition method.
出处
《光电子．激光》
EI
CAS
CSCD
北大核心
2010年第4期588-592,共5页
Journal of Optoelectronics·Laser
基金
教育部"春晖计划"科研合作资助项目(Z2005-2-11009)
辽宁省创新团队资助项目(2006T102)
