An Efficient Feature Extraction Method Based on Kernel Maximum Margin Criterion

摘要：

Based on kernel maximum margin criterion (KMMC), a new method for nonlinear feature extraction is presented in this paper. Compared with original KMMC feature extraction method and kernel principal component analysis (KPCA) method, the proposed method has more powerful capability to eliminate the statistical correlation between features and improve efficiency of feature extraction. Our experimental results show that our new method is better than original KMMC and KPCA in terms of efficiency and stability on Olivetti Research Laboratory (ORL) face database by leave-one-out method.

作者: Yong-Zhi Li Jing-Yu Yang Guo-Dong Li Zhao-Cheng Qiu

作者单位: Department of Computer Science Nanjing University of Science & Technology.Nanjing 210094, China;Scho Department of Computer Science Nanjing University of Science & Technology.Nanjing 210094, China, School of Information Engineering,Lanzhou Commercial College,Lanzhou 730020, China Department of Mathematics,East China Normal University.Shanghai 200062, China,

会议类型: 国际会议

会议名称: 第三届国际脉冲动力系统及应用学术会议

会议地点: 青岛

会议语种:英文

在线出版日期: 2006-07-21（万方平台首次上网日期，不代表论文的发表时间）