Improving Spectroface using Pre-processing and Voting

摘要：

In this paper we present pre-processing steps and a voting scheme that improve the effectiveness of the spectroface approach. It consists on a series of preprocessing steps prior to spectroface togheter with a texture feature that are used independently. The classifier output for each of the 13 features is fused using a majority voting scheme coupled with rules for ties and strong features. Yale (15 subjects), Olivetti (40 subjects) and Notre Dame (487 subjects) face databases are selected to evaluate the proposed method yielding 97.33, 85.28 and 75.91％ accuracy, respectively, using only one training image per subject.

作者: Ricardo Santos Lufs A.Alexandre

作者单位: Dept.Informatics,University of Beira Interior,Portugal Dept.Informatics,University of Beira Interior and Instituto de Telecomunicacoes Covilha,Portugal

会议类型: 国际会议

会议名称: 2009 IEEE International Conference on Intelligent Computing and Intelligent Systems(2009 IEEE 智能计算与智能系统国际会议)

会议地点: 上海

会议语种:英文

页码: 3081-3085

在线出版日期: 2009-11-20（万方平台首次上网日期，不代表论文的发表时间）