東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Biometric identity verification base...

Shen, Tsu-Wang.

Linked to FindBook

Google Book

Amazon

博客來

Biometric identity verification based on electrocardiogram (ECG).

Record Type:	Electronic resources : Monograph/item
Title/Author:	Biometric identity verification based on electrocardiogram (ECG)./
Author:	Shen, Tsu-Wang.
Description:	166 p.
Notes:	Source: Dissertation Abstracts International, Volume: 66-05, Section: B, page: 2695.
Contained By:	Dissertation Abstracts International66-05B.
Subject:	Engineering, Biomedical. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3175469
ISBN:	9780542139826

Biometric identity verification based on electrocardiogram (ECG).
Shen, Tsu-Wang.

Biometric identity verification based on electrocardiogram (ECG). - 166 p.

Source: Dissertation Abstracts International, Volume: 66-05, Section: B, page: 2695.

Thesis (Ph.D.)--The University of Wisconsin - Madison, 2005.

The electrocardiogram (ECG), a new biometric for human identification, is investigated in this research. Selected features extracted from lead-I electrocardiogram (ECG) are used to identify a person in a predetermined group. This research focused on short-term, resting, and lead-I palm ECG signals. The single-lead ECG is a one-dimensional, low-frequency signal that can be recorded easily from two hands.

ISBN: 9780542139826Subjects--Topical Terms:

1017684
Engineering, Biomedical.

Biometric identity verification based on electrocardiogram (ECG).
LDR:02994nmm 2200289 4500 001 1829554
005 20061114131355.5
008 130610s2005 eng d
020 $a 9780542139826
035 $a (UnM)AAI3175469
035 $a AAI3175469
040 $a UnM $c UnM
100 1 $a Shen, Tsu-Wang. $3 1918412
245 1 0 $a Biometric identity verification based on electrocardiogram (ECG).
300 $a 166 p.
500 $a Source: Dissertation Abstracts International, Volume: 66-05, Section: B, page: 2695.
500 $a Supervisor: Willis J. Tompkins.
502 $a Thesis (Ph.D.)--The University of Wisconsin - Madison, 2005.
520 $a The electrocardiogram (ECG), a new biometric for human identification, is investigated in this research. Selected features extracted from lead-I electrocardiogram (ECG) are used to identify a person in a predetermined group. This research focused on short-term, resting, and lead-I palm ECG signals. The single-lead ECG is a one-dimensional, low-frequency signal that can be recorded easily from two hands.
520 $a The purpose of this research is to analyze ECG features, to design a human identify verification system, to provide an efficient feature selection method, and to evaluate performance of the identify verification system. There are three phases to this study. (1) The first phase was to determine if a one-lead ECG is a potentially applicable method for human identification. One-lead ECG signals of 20 individuals from the MIT/BIH (clinical) database were investigated. Mainly, template matching and decision-based neural network (DBNN) are discussed in this phase. The experimental results of using these two methods separately showed the rate of correct identification to be up to 95% in a predetermined group of subjects. Also, significantly, if the two methods are combined together, a 100% correct rate was achieved. (2) The second phase was to verify if biometric ECG technology can be applied for normal, healthy people. Short-term, resting, Lead-I ECG signals were measured from 168 individuals to create our ECG biometric database. Multiple algorithms were evaluated for ECG identification. The algorithms included template matching, mean-square-error matching, distance classification, and DBNN methods. (3) The third phase was to develop a feature selection method and to test the method using four different databases. The computer simulation results show that our feature selection method efficiently and successfully increased the system performance in four different databases and several classifiers. Also, the selected ECG features increased our biometric system performance from 84.52% to 95.3% accuracy in a large group with 168 normal healthy people.
590 $a School code: 0262.
650 4 $a Engineering, Biomedical. $3 1017684
650 4 $a Biology, Biostatistics. $3 1018416
690 $a 0541
690 $a 0308
710 2 0 $a The University of Wisconsin - Madison. $3 626640
773 0 $t Dissertation Abstracts International $g 66-05B.
790 1 0 $a Tompkins, Willis J., $e advisor
790 $a 0262
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3175469