摘要
提出了一种基于局部二元模式(Local Binary Pattern,LBP)与支持向量机(SVM)相结合的面部表情识别方法。使用LBP算子对图像进行处理,对图像的模式进行统计形成面部表情特征;使用线性判别分析对表情特征进行降维处理;采用支持向量机对面部表情进行分类。用Matlab实现了上述方法,并在日本女性人脸表情(JAFFE)数据库上测试,取得了70.95%的识别率。
A novel approach to facial expression recognition based on the combination of Local Binary Pattern(LBP) and Support Vector Machine(SVM) is proposed.First,the algorithm processes facial expression images with LBP operator and then facial expression features are formed by statistics of image's LBPs.Then Linear Discrimination Analysis(LDA) method is used for feature di- mension reduction and SVM for classification.Finally,the algorithm is implemented with Matlab and experimented in Japanese Female Facial Expression database(JAFFE database ).A recognition rate of 70.95% is obtained and shows the effectiveness of the proposed algorithm.
出处
《计算机工程与应用》
CSCD
北大核心
2009年第29期180-183,共4页
Computer Engineering and Applications
基金
广东省自然科学基金(No.07010869)
北京大学视觉与听觉信息处理国家重点实验室开放课题基金项目(No.0505)
浙江大学CAD&CG国家重点实验室开放课题(No.A0703)~~
关键词
面部表情识别
局部二元模式
线性判别分析
支持向量机
facial expression recognition
Local Binary Pattern (LBP)
Linear Discrimination Analysis (LDA)
Support Vector Machine(SVM)