| 低分辨率人脸图像识别性能研究 |
| |
| 引用本文： | 林忠.低分辨率人脸图像识别性能研究[J].南京工程学院学报(自然科学版),2009,7(4):42-46. |
| |
| 作者姓名： | 林忠 |
| |
| 作者单位： | 南京工程学院计算机工程学院,江苏,南京,211167 |
| |
| 摘 要： | 首先介绍了人脸识别中常用的PCA方法、LDA方法和Fisherface方法，然后通过在ORL数据库和Yale数据库上的试验发现当分辨率从原始分辨率降低到20×20左右的过程中，原始空间、PCA子空间和Fisherface子空间方法的识外浊能均没有显著变化．降低分辨李能有效提高现有人脸识剐算法的速度．
|
| 关 键 词： | PCA LDA Fisherface 低分辨率 |
Research of Low-resolution Face Recognition Performance |
| |
| LIN Zhong.Research of Low-resolution Face Recognition Performance[J].Journal of Nanjing Institute of Technology :Natural Science Edition,2009,7(4):42-46. |
| |
| Authors: | LIN Zhong |
| |
| Institution: | LIN Zhong ( School of Computer Engineering, Nanjing Institute of Technology, Nanjing 211167, China) |
| |
| Abstract: | When it comes to face recognition, little research into the influence of image resolution on recognition performance has been conducted so far. This paper introduces some face recognition methods widely used in practice, for instance, PCA, LDA and Fisherface. The experiments carried out with ORL data and Yale data indicate that no dramatic changes take place in recognition performance of the original space, PCA sub-space and Fisherface sub-space when the resolution decreases to 20 plus 20. It shows that decrease in resolution can improve the current face recognition arithmetic. |
| |
| Keywords: | PCA LDA Fisherface |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
