|Table of Contents|

Cost一sensitive Sparse Locality Preserving Projections(PDF)

《哈尔滨理工大学学报》[ISSN:1007-2683/CN:23-1404/N]

Issue:
2015年03期
Page:
45-050
Research Field:
计算机与控制工程
Publishing date:

Info

Title:
Cost一sensitive Sparse Locality Preserving Projections
Author(s):
LIN Ke一ZhengZHONG Yan
CHENG Wei-yue
(School of Computer Science and Technology,Harbin University of Science and Technology, Harbin 150080,China
Keywords:
locality preserving projectionsface recognitioncost-sensitivesparse
PACS:
TP391 .4
DOI:
-
Abstract:
In Locality Preserving Projections algorithm,faces in similar categories are projected as the sameone,leading to the decrease of recognition rate. To solve this problem,Cost-sensitive Sparse Locality PreservingProjections algorithmbased on LPP algorithm is proposed. In CSLPP algorithm,in which Cost一Sensitive Learningwas applied to face recognition,face samples were first cost-sensitively thought of, and Sparseness,at last the opti-mal projection vector was figured out. Experimental results on the YALE and FERET face databases show that CLPP algorithm effectively avoids high risks and its recognition rate is significantly higher than that of others in Nearest Neighbor Classifier.

References:

-

Memo

Memo:
-
Last Update: 2015-08-24