(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Biometric identity verification system based on fingerprint detection and mykad

Biometric identity verification system based on fingerprint detection and mykad / Victor Teoh De Zhi
Pengesahan identiti adalah sangat penting dalam dunia zaman kini dan ia digunakan dalam kehidupan harian kita. Untuk pelajar, kita mempunyai kad metrik dan untuk pekerja, mereka mempunyai kad pintar tersendiri sebagai alat pengesahan identiti. Namun, sistem pengesahan identiti ini mempunyai kelemahan dan ia boleh disalah gunakan oleh pihak yang tidak bertanggungjawab. Integrasi sistem pengesahan melalui cap jari dan MyKad dapat mengetatkan mengatasi kelemahan ini kerana cap jari setiap orang adalah unik dan setiap rakyat Malaysia hanya mempunyai sekeping MyKad. Objektif projek ini adalah untuk mencipta sebuah sistem pengesahan identiti yang kukuh, pantas dan senang untuk diguna. Cap jari seseorang mempunyai banyak maklumat yang terperinci dan maklumat ini disimpan sebagai Kod Jari dalam sistem ini. Lokasi teras cap jari dikesan melalui penapis kompleks tahap 1 dan kaedah ambang Otsu. Kawasan teras cap jari diasingkan dan dinormalisasikan untuk mendapatkan imej yang baik. Teras cap jari diasingkan kepada 64 bahagian dalam bulatan. Selepas itu, cap jari tersebut diproses dengan penapis Gabor 8-arah untuk mendapatkan aliran rabung dan kerut. Setiap cap jari yang telah diproses disimpan sebagai Kod Jari. Proses pengesahan cap jari dilakukan dengan mengira jarak Euclidean di antara Kod Jari yang dikesan dengan Kod Jari di pangkalan data. Jika jarak tersebut lebih daripada ambang yang diletakkan, cap jari tersebut adalah disahkan. Sistem ini telah berjaya mendapat keputusan pengesahan cap jari setinggi 90% dari 40 percubaan daripada 10 orang yang berbeza. Sebanyak 100 cap jari daripada 10 orang telah disimpan sebagai Kod Jari dalam pangkalan data. Proses pengesahan untuk MyKad mengambil masa 11.02 saat dan 1.3 saat untuk pengesahan cap jari. _______________________________________________________________________________________________________ Identity recognition is very important in the world today and it is used in our everyday life. For students, we have student matric cards and for workers or employees, they have their own smart card to scan in and out during their work. However, this recognition system has a weakness and it might be violated and abused. The integration between fingerprint verification and MyKad verification elevates the security in identification to the highest level as the fingerprints are unique from one another and there is only 1 MyKad for each citizen in Malaysia. The objectives for this project is to create a user friendly and secure identity verification system with fast matching algorithm. Fingerprints contain a lot of minute information and they are called minutiae. These minutiae are captured and stored inside Fingercodes. The core of the fingerprints are detected using level 1 complex filtering with Otsu’s threshold method to get the region of interest. The core of the fingerprint is cropped and normalized to avoid unwanted noise that could affect the accuracy of the system. Tessellation is the method used to segment the core to 64 sectors. After that, the cropped fingerprint core is convolved with 8-directions of Gabor filters to get the flow of the ridges and furrows. Each Gabor filtered result is saved as a Fingercode. The verification process calculates the Euclidean distance of the detected Fingercode with the Fingercode in the database and if the distance is more than an empirically defined threshold value, the fingerprint is verified. The verification success rate was 90% out of 40 attempts from 10 different people. There are a total of 100 fingerprints from 10 people which is converted into Fingercode in the database. The processing time for MyKad verification is 11.02s for the full mode and 0.62seconds for the “lite" mode. The average processing time for the fingerprint verification is 1.3seconds.
Contributor(s):
Victor Teoh De Zhi - Author
Primary Item Type:
Final Year Project
Identifiers:
Accession Number : 875006274
Language:
English
Subject Keywords:
Identity; recognition; fingerprint
First presented to the public:
6/1/2014
Original Publication Date:
8/26/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 96
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-08-26 09:43:27.718
Submitter:
Mohd Jasnizam Mohd Salleh

All Versions

Thumbnail Name Version Created Date
Biometric identity verification system based on fingerprint detection and mykad1 2020-08-26 09:43:27.718