Please use this identifier to cite or link to this item:
Title: Application of a facial and palmprint based multimodal biometric system
Authors: Kamel, Antoine
Advisors: Daba, Jihad S. 
Subjects: Biometrics (Biology)
Optical data processing
Pattern recognition
Issue Date: 2020
Biometric systems have become way popular these days; A biometric trait such as a fingerprint is a lot more unique, portable and secured than a password or a physical card, you just cant forget or lose your fingerprint, nor it can be easily stolen or forged. Now moreover, a biometric system relying on multiple biometric traits (Multimodal) guarantees more security and is much more accurate and efficient than single-trait systems (Unimodal); It will take advantage of this combination of traits to overcome the limitations that single-trait systems pose. For instance, fingerprints might deteriorate over time with age, so the presence of Face Recognition alongside Fingerprint Recognition in one system can give the user the option to use one of the two or even both at the same time. We introduce in this paper a multimodal biometric system that combines both facial and palm recognition (traits) in order to authenticate the identity of the input user; We detail each part forming the system, starting from obtaining inputs, to region of interest (ROI) extraction, processing of each ROI, comparing results to the databases, generating scores then fusing them into one score in order to decide whether to authenticate or no.
Includes bibliographical references (p. 31).

Supervised by Dr. Jihad Daba.
Rights: This object is protected by copyright, and is made available here for research and educational purposes. Permission to reuse, publish, or reproduce the object beyond the personal and educational use exceptions must be obtained from the copyright holder
Ezproxy URL: Link to full text
Type: Project
Appears in Collections:UOB Theses and Projects

Show full item record

Record view(s)

checked on Oct 24, 2021

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.