面向人脸表情识别的双模板稀疏分类方法

提出一种面向人脸表情识别的双模板稀疏分类方法（DT—SRC）。该算法在用训练样本组成观测矩阵的基础上，通过添加正、负双模板构造新的观测矩阵，最后使用稀疏表示分类（SRC）进行识别。分别在JAFFE和CK人脸库中进行验证，结果表明，该算法识别准确率高，比SRC有更好的性能。...

Full description

Saved in:
Bibliographic Details
Published in电子技术应用 Vol. 40; no. 6; pp. 119 - 122
Main Author 蒋行国 冯彬 韦保林
Format Journal Article
LanguageChinese
Published 桂林电子科技大学通信与信息学院,广西桂林,541004 2014
Subjects
双模板
稀疏表示
表情识别
观测矩阵
measurement matrix
dual templates
表情识别
观测矩阵
双模板
facial expression recognition
稀疏表示
sparse representation
Online AccessGet full text
ISSN0258-7998

Cover

More Information
Summary:提出一种面向人脸表情识别的双模板稀疏分类方法（DT—SRC）。该算法在用训练样本组成观测矩阵的基础上，通过添加正、负双模板构造新的观测矩阵，最后使用稀疏表示分类（SRC）进行识别。分别在JAFFE和CK人脸库中进行验证，结果表明，该算法识别准确率高，比SRC有更好的性能。
Bibliography:facial expression recognition; sparse representation; measurement matrix; dual templates
11-2305/TN
This paper proposes dual templates and sparse classification algorithm（DT-SRC） for facial expression recognition. The algorithm adds positive and negative dual templates to construct the new observation matrix on the basis of using the training sam- pie for observation matrix, finally using SRC for identification. Respectively verification in JAFFE and CK face shows that the algorithm has high recognition accuracy rate, and the performance is better than the SRC.
Jiang Xingguo, Feng Bin, Wei Baolin （Guilin University of Electronic Technology, School of Information and Communication, Guilin 541004, China）
ISSN:0258-7998