A Secure Personal Identification System: Techniques of Fingerprint Image Enhancement, Feature Extraction and Template Protection

Citation

Islam, Md. Rajibul (2010) A Secure Personal Identification System: Techniques of Fingerprint Image Enhancement, Feature Extraction and Template Protection. LAP LAMBERT Academic Publishing AG & Co. KG, Germany. ISBN 978-3-8383-7028-6

[img] Image (JPEG)
Cover_Image.jpg - Cover Image

Download (0B)

Abstract

Biometrics, which is sometimes called biometric recognition, refers to the recognition of individuals based on their physiological (e.g., fingerprints, face, retina, iris) and/or behavioral (e.g., gait, signature) characteristics. In this book, an explanation is given of the usage of fingerprints from an integrity point of view, how to improve the captured images by enhancement and feature extraction when it is a poor and low resolution fingerprint image, and how to protect templates in order to enhance the security and privacy issue of the entire authentication system. Some problems with current systems have been identified. In order to solve these difficulties, this study has been divided into three parts. In the first part, an improved preprocessing approach is presented using gamma manipulation and gamma correction techniques to adjust lightness and intensities of the poor image due to enhancing the quality. The second part is to improve the feature extraction technique using Projection Incorporated Subspace based on PCA. The third part is a protection analysis of foremost privacy enhanced technologies including watermark embedding technique with fixed digit encryption.

Item Type: Book
Subjects: Q Science > Q Science (General)
Q Science > QA Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Divisions: Faculty of Information Science and Technology (FIST)
Depositing User: Users 28 not found.
Date Deposited: 05 Jul 2010 08:44
Last Modified: 05 Jul 2010 08:44
URII: http://shdl.mmu.edu.my/id/eprint/836

Downloads

Downloads per month over past year

View ItemEdit (login required)