作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (14): 167-168. doi: 10.3969/j.issn.1000-3428.2011.14.055

• 人工智能及识别技术 • 上一篇    下一篇

稀疏保留判决分析在人脸表情识别中的应用

黄 勇   

  1. (柳州铁道职业技术学院电子工程系,广西 柳州 545007)
  • 收稿日期:2010-12-22 出版日期:2011-07-20 发布日期:2011-07-20
  • 作者简介:黄 勇(1980-),男,讲师、硕士,主研方向:模式识别,图像处理,自动检测技术

Application of Sparse Preserving Discriminant Analysis in Facial Expression Recognition

HUANG Yong   

  1. (Department of Electronic Engineering, Liuzhou Railway Vocational Technical College, Liuzhou 545007, China)
  • Received:2010-12-22 Online:2011-07-20 Published:2011-07-20

摘要: 提出一种基于稀疏保留判决分析的人脸表情识别方法——SPDA方法。引入稀疏描述理论结合半监督判决分析SDA,通过稀疏重构处理,可获得图像的局部结构信息。由于稀疏描述本身具有的判决性,SPDA只需少量的样本就能获得较好的效果。CED-WYU和JAFFE的2个表情数据库的识别结果表明,该方法能有效提高识别率。

关键词: 数据降维, 线性判决分析, 半监督判决分析, 稀疏保留判决分析, 人脸表情识别

Abstract: A facial expression recognition method based on Sparse Preserving Discriminant Analysis(SPDA) is proposed. The graph in SPDA is constructed by sparse representation, and thus the local structure information is automatically modeled, and with the natural discriminative power of sparse representation, SPDA can get better performance only resorting to very few extra unlabeled samples. Experimental result on CED-WYU and JAFFE show that SPDA is an effective method for improving the recognition accuracy.

Key words: dimensionality reduction, Linear Discriminant Analysis(LDA), Semi-supervised Discriminant Analysis (SDA), Sparse Preserving Discriminant Analysis(SPDA), facial expression recognition

中图分类号: