(For USM Staff/Student Only)

EngLib USM > Ω School of Electrical & Electronic Engineering >

Image quality measurement for fingerprint images/Saw Beng Keat

Image quality measurement for fingerprint images_Saw Beng Keat_E3_2013_NI
Pada masa kini, pengenalan cap jari adalah salah satu teknik biometrik yang paling popular dan ia digunakan secara meluas dalam kebanyakan aplikasi. Walau bagaimanapun, prestasi sistem pengenalan bergantung pada kualiti imej cap jari secara kritikal. Imej-imej berkualiti rendah akan menyebabkan ciri-ciri hilang atau palsu, yang menyebabkan kemerosotan prestasi sistem pengenalan tersebut. Oleh itu, penilaian kualiti imej cap jari sebelum padanan adalah penting. Dalam projek ini, satu kaedah yang telah dicadang berdasarkan Principal Component Analysis (PCA) telah dipilih untuk dilaksanakan di komputer peribadi dan platform Android. PCA merupakan satu kaedah statistik yang umum dan berguna untuk mengenalpasti corak dari data yang berdimensi tinggi. Dalam kaedah PCA, dua ciri telah dipetik untuk menjana ukuran kualiti tempatan dari setiap blok imej. Kemudian, dengan menggabungkan Harris-Corner Strength (HCS) dinormalkan sebagai nilai wajaran untuk ukuran kualiti blok tempatan, kualiti global imej cap jari boleh dianggarkan. Algoritma yang sama telah dilaksanakan pada komputer dan platform Android. Kaedah yang digunakan juga dinilai berdasarkan pangkalan data cap jari pendam dari Indraprastha Institute of Information Technology, Delhi (IIIT-D). Keputusan yang diperoleh akan dibandingkan dengan keputusan perisian NIST Fingerprint Image Quality (NFIQ). Perbandingan menunjukkan bahawa sistem yang dibangunkan tidak dapat mengklasifikasikan kualiti cap jari secara betul. Ia mungkin disebabkan oleh aspek-aspek berbeza yang digunakan untuk meneliti kualiti oleh keduadua aplikasi. ___________________________________________________________________________________ Nowadays, fingerprint identification is one of the most popular biometric techniques and it is widely used in many applications. However, the performance of identification system relies critically on the quality of the fingerprint images. Poor quality images will result in missing or fake features, causing degradation of the performance of the identification system. Therefore, it is important to estimate the quality of the fingerprint images before matching. In this project, a proposed method based on Principal Component Analysis (PCA) is selected to implement in personal computer and Android platforms. PCA is a common and useful statistical method for identifying patterns from high-dimension data. In PCA method, two novel features are extracted to generate a local quality measure from each image block. Then, by combining the normalized Harriscorner strength (HCS) as weighted value into local block quality measure, the global quality of the fingerprint image can be estimated. The same algorithm has been implemented on both computer and Android platforms. The method used has also been evaluated on the latent fingerprint database from Indraprastha Institute of Information Technology, Delhi (IIIT-D). The results obtained are compared with the results of NIST Fingerprint Image Quality (NFIQ) Software. The comparison shows that the developed system is unable to classify the quality of fingerprint correctly. It may be due to totally different aspects used to examine the quality.
Contributor(s):
Saw Beng Keat - Author
Primary Item Type:
Final Year Project
Language:
English
Subject Keywords:
fingerprint ; biometric ; images
First presented to the public:
6/1/2013
Original Publication Date:
2/18/2020
Previously Published By:
Universiti Sains Malaysia
Place Of Publication:
School of Electrical & Electronic Engineering
Citation:
Extents:
Number of Pages - 97
License Grantor / Date Granted:
  / ( View License )
Date Deposited
2020-02-19 10:38:29.905
Submitter:
Nor Hayati Ismail

All Versions

Thumbnail Name Version Created Date
Image quality measurement for fingerprint images/Saw Beng Keat1 2020-02-19 10:38:29.905