ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper Biometric Authentication Based on Feature Extraction of Bioelectric Transfer Characteristics of Fingers
Cited - time in scopus Share share facebook twitter linkedin kakaostory
Authors
Hyung Wook Noh, Joo Yong Sim, Bong Kyu Kim, Chang Geun Ahn
Issue Date
2018-07
Citation
International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) 2018, pp.1-1
Publisher
IEEE
Language
English
Type
Conference Paper
Abstract
Recent studies reveal the possibility of spoofing biometric systems based on acquired image by using artificial biometric samples such as fingerprints etc. Therefore, we suggest new method of bioelectric finger impedance based on feature extraction of signal with frequency domain. 73 samples of finger impedance were acquired among 12 individuals. Linear support vector machine was used to train the measured data, and we achieved 98.6% of accuracy by using 5-fold cross validation.
KSP Keywords
5-fold cross-validation, Biometric authentication, Cross validation(CV), Feature extractioN, Frequency domain(FD), Linear support vector machine(SVM), biometric system, measured data, new method, transfer characteristics, vector machine(LSSVM)