Reduced set kernel principal component analysis (rskpca) algorithm for palm print based mobile biometric system
Loading...
Date
2015-05-01
Authors
Noor Salwani Ibrahim
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The emerging of internet and wireless dimension has brought a new era in biometrics
technology. Instead of operating the biometric system with static biometric device,
mobile biometric system can be implemented and this approach leads to more
efficient and reliable implementation. In this study mobile biometric system based on
palm print modality is developed. However, in order to execute mobile biometric
system, efficient processing time and storage are some of the important factors that
need to be considered. In this research, algorithms involving palm print feature
processing are evaluated so as to obtain optimum time and memory consumption.
Several feature processing methods including Region of Interest (ROI), Principal
Component Analysis (PCA), and Kernel Principal Component Analysis (KPCA) are
investigated. A new approach in feature extraction called Reduced-Set Kernel
Principal Component Analysis (RSKPCA) is proposed to speed up the processing in
feature extraction. The proposed RSKPCA employs a Reduced Set Density Estimate
(RSDE) to define a weighted gram matrix. As a result, the RSKPCA only extracts
the most relevant and important information from a dataset. 2400 palm print images
which were collected from three types of android mobile are employed.
Experimental evaluation shows that the proposed RSKPCA has better performance
compared to the ROI, PCA and KPCA with the Genuine Acceptance Rates (GAR) is
more than 98% and the matching time is less than 0.5s. In this project, it has been
proven that the proposed RSKPCA as feature extraction gives the best result for
mobile biometric system based on palm print.