Paper Info
Title
Two-directional two-dimensional random projection and its variations for face and palmprint recognition
Abstract
2DRP (two-dimensional random projection) is two-dimensional extension of one-dimensional RP (random projection) to keep biometric images from being reshaped to vectors before RP for recognition. We propose a novel method called (2D)2RP (two-directional two-dimensional random projection) for feature extraction of biometrics. (2D)2RP directly projects the image matrix from high-dimensional space to low-dimensional space to extract optimal projective vectors at row-direction and column-direction. (2D)2RP, similar to RP, can also avoid the problems of singularity, SSS (small sample size) and over-fitting; furthermore it has much less storage and computational cost than RP. Besides, the variations of (2D)2RP combined with 2DPCA and 2DLDA are developed. Experimental results and comparison discussion among (2D)2RP and its variations on face and palmprint databases confirm the performance and effectiveness of (2D)2RP and its variations.
Year
DOI
Venue
2011
10.1007/978-3-642-21934-4_37
ICCSA (5)
Keywords
Field
DocType
two-directional two-dimensional random projection,two-dimensional random projection,palmprint recognition,two-dimensional extension,one-dimensional rp,high-dimensional space,computational cost,random projection,comparison discussion,biometric image,linear discriminant analysis,face recognition,principal component analysis
Random projection,Computer vision,Facial recognition system,Pattern recognition,Computer science,Matrix (mathematics),Singularity,Feature extraction,Artificial intelligence,Linear discriminant analysis,Biometrics,Principal component analysis
Conference
Volume
ISSN
Citations 
6786
0302-9743
17
PageRank 
References 
Authors
0.52
9
5
Name
Order
Citations
PageRank
Lu Leng12009.83
Jiashu Zhang2112275.03
Gao Chen3534.78
Muhammad Khurram Khan43538204.81
Khaled Alghathbar549832.54