Jannah, Raudhatul (2015) Identifikasi Personal Biometrik Berdasarkan Sinyal Photoplethysmography dari Detak Jantung. Masters thesis, Institute Tehcnology Sepuluh Nopember.
Preview |
Text
2213205025-Master_Thesis.pdf - Accepted Version Download (2MB) | Preview |
Abstract
Sistem biometrik sangat berguna untuk membedakan karakteristik individu
seseorang. Sistem identifikasi yang paling banyak digunakan diantaranya berdasarkan
metode fingerprint, face detection, iris atu hand geometry. Penelitian ini mencoba
untuk meningkatkan sistem biometrik menggunakan sinyal Photoplethysmography
dari detak jantung. Algoritma yang diusulkan menggunakan seluruh ektraksi fitur
yang didapatkan melalui sistem untuk pengenalan biometrik. Efesiensi dari algoritma
yang diusulkan didemonstrasikan oleh hasil percobaan yang didapatkan menggunakan
metode klasifikasi Multilayer Perceptron, Naïve Bayes dan Random Forest
berdasarkan fitur ekstraksi yang didapatkan dari proses sinyal prosesing. Didapatkan
51 subjek pada penelitian ini; sinyal PPG signals dari setiap individu didapatkan
melalui sensor pada dua rentang waktu yang berbeda. 30 fitur karakteristik didapatkan
dari setiap periode dan kemudian digunakan untuk proses klasifikasi. Sistem
klasifikasi menggunakan metode Multilayer Perceptron, Naïve Bayes dan Random
Forest; nilai true positive dari masing-masing metode adalah 94.6078 %, 92.1569 %
dan 90.3922 %. Hasil yang didapatkan menunjukkan bahwa seluruh algoritma yang
diusulkan dan sistem identifikasi biometrik dari pengembangan sinyal PPG ini sangat menjanjikan untuk sistem pengenalan individu manusia.
=============================================================================================
The importance of biometric system can distinguish the uniqueness of personal
characteristics. The most popular identification systems have concerned the method
based on fingerprint, face detection, iris or hand geometry. This study is trying to
improve the biometric system using Photoplethysmography signal by heart rate. The
proposed algorithm calculates the contribution of all extracted features to biometric
recognition. The efficiency of the proposed algorithms is demonstrated by the
experiment results obtained from the Multilayer Perceptron, Naïve Bayes and
Random Forest classifier applications based on the extracted features. There are fifty
one persons joined for the experiments; the PPG signals of each person were recorded
for two different time spans. 30 characteristic features were extracted for each period
and these characteristic features are used for the purpose of classification. The results
were evaluated via the Multilayer Perceptron, Naïve Bayes and Random Forest
classifier models; the true positive rates are then 94.6078 %, 92.1569 % and 90.3922
%, respectively. The obtained results showed that both the proposed algorithm and the
biometric identification model based on this developed PPG signal are very promising
for contact less recognizing systems.
| Item Type: | Thesis (Masters) |
|---|---|
| Additional Information: | RTE 006.4 Jan b |
| Uncontrolled Keywords: | PPG, Heart rate, Algorithms |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.B56 Biometric identification |
| Divisions: | Faculty of Industrial Technology > Electrical Engineering > 20101-(S2) Master Thesis |
| Depositing User: | Yeni Anita Gonti |
| Date Deposited: | 16 Mar 2020 15:39 |
| Last Modified: | 16 Mar 2020 15:39 |
| URI: | http://repository.its.ac.id/id/eprint/75491 |
Actions (login required)
 |
View Item |