Ear Recognition based on Wavelet Transform and Discriminative Common Vectors

摘要：

Discriminative Common Vectors is one of the mostsuccessful methods which overcome the small samplesize case in Fishers Linear Discriminant Analysis.Butwhen we directly use DCV to reduce the dimensions ofthe ear images,the computational expense of trainingis still relatively large.A new method is proposed inthis paper that the low frequency sub-images areobtained by utilizing two-dimensional wavelettransform firstly and the features are extracted byapplying DCV to the sub-images.The experimentalresults show that the proposed Wavelet+ DCV methodachieve better performance than PCA+LDA methodand the computation burden is also reduced greatly.

作者: Hai-long Zhao Zhi-chun Mu Xia Zhang Wen-jie Dun

作者单位: University of Science and Technology Beijing,Beijing 100083,China

会议类型: 国际会议

会议名称: 2008 3rd International Conference on Intelligent System and Knowledge Engineering(第三届智能系统与知识工程国际会议)(ISKE 2008)

会议地点: 厦门

会议语种:英文

页码: 713-716

在线出版日期: 2008-11-17（万方平台首次上网日期，不代表论文的发表时间）