摘要
将基于最大间距准则的人脸特征提取方法进行改进,以提高人脸识别率。通过在类间散度矩阵中增加距离调节函数,经选择得到最佳鉴别矢量,用支持向量机的多分类算法对改进准则所提取的特征进行识别。在ORL人脸数据库上的试验结果表明,改进算法对人脸光照和姿态变化具有较强的鲁棒性,较传统算法在特征提取方面更加有效。
In order to increase face recognition rate,improves the facial feature extraction method of maximum margin criterion.Through increasing the distance tuning function into the between-class scatter matrix,selects the optimal discriminant vectors.Uses support vector machine multi-classification algorithm to identiy the extracted features of the improved methods.The experiment result on ORL face database shows that the improved method is robust to illumination and pose,and it is better than the traditional method in terms of efficiency about feature extraction.
出处
《现代计算机》
2010年第9期51-54,共4页
Modern Computer
关键词
人脸识别
最大间距准则
支持向量机
Face Recognition
Maximum Margin Criterion
Support Vector Machine
