-
摘要: 人脸识别技术是模式识别和机器视觉领域的一个重要研究方向，在众多人脸识别的算法中，基于子空间分析的特征提取方法以其稳定可靠的识别效果成为了人脸识别中特征提取的主流方法之一。本文对目前应用较多的子空间分析方法进行了研究，具体介绍了线性子空间分析方法：主成分分析（PCA）、线性鉴别分析（LDA）、独立主成分分析（ICA）、快速主成分分析（FastICA）等及非线性子空间分析方法：基于核的PCA （KPCA）等的基本思想及其在人脸识别中的研究进展，包括一些新的研究成果。此外，还应用orl及Yale B人脸库对几个基础的子空间方法进行了验证实验。实验结果表明，在几个子空间分析方法中，FastICA算法取得了最高的识别率。最后结合实验结果对各算法的优缺点进行了分析总结。Abstract: Face recognition is an important problem of pattern recognition and machine learning. Among many approaches to the problem of face recognition, subspace analysis gives the most promising results, and becomes one of the most popular methods. This paper researches subspace analysis methods, introduces the basic theory of linear subspace such as PCA、LDA、ICA 、FastICA etc. and non-linear subspace such as KPCA etc. and their application in face recognition ,including some new research fruits concretely . In addition ,ORL database and YALE B database are used to experiment basic subspace methods. The experiment results indicate that FastICA method is more powerful than other subspace methods for face recognition. Finally, the advantage and disadvantage of these methods are discussed by the experiment results.
-
点击查看大图
计量
- 文章访问数: 4145
- HTML全文浏览量: 191
- PDF下载量: 4120
- 被引次数: 0
下载:
