Document Details

Document Type : Thesis 
Document Title :
Using Biometric Techniques in Healthcare Security Systems
استخدام تقنيات القياسات الحيوية في نظم سرية الرعاية الصحية
 
Subject : Faculty of Computing and Information Technology 
Document Language : Arabic 
Abstract : Security is considered a corner stone for health-care information systems as they contain extremely sensitive information. The aim is to provide healthcare personnel access to the right information at the right time while ensuring high patient privacy. Biometrics play an important role in healthcare applications, especially when there is a need to control access through identification of authorized users. Face and Iris biometrics have been used separately for access security. Although face recognition is user friendly and non-invasive, it has low distinctiveness. On the other hand, iris recognition is one of the most accurate biometrics, but must meet stringent quality criteria. The significance of fusing these two biometrics is more than the improvement in identification accuracy. Enlarging user population coverage and reducing enrolment failure are additional reasons for combining face and iris for identification. In this thesis a hierarchical architecture of Electronic Medical Record (EMR) multimodal biometric identification system has been proposed based on sensitivity information security level. The aim is to apply the proposed multimodal biometric identification system to the highest level of security in EMR. In order to improve the authentication performance and reduce the spoof attacks, we have built multimodal biometric identification system that combines information from both face and iris unimodal. After suitable normalization of scores, fusion is performed at the match score level using matcher weighting scheme. The effect of changing the number and quality of images is tested on four combination sets (CS1-CS4). The system performance is evaluated on the Olivetti Research Laboratory (ORL) face database and Chinese Academy of Sciences: Institute of Automation (CASIA) version-1 iris database. The study found that increasing the number of images alone without including various image qualities in unimodal system will not insure performance improvement. By this we mean, the overall group of images used should be composed of different image quality to match the actual circumstances of the real user. The overall system performance improved with the proposed multimodal biometric identification system. The Recognition rate of the proposed multimodal biometric identification system improved by 31% and 1% over face and iris unimodal, respectively. The study concluded that the combination of face and iris unimodal into our proposed multimodal biometric identification system has higher performance than each unimodal separately. We observed that multimodal biometric is a way to reduce the quality requirement of images and makes spoofing attack much more difficult task. 
Supervisor : ر 
Thesis Type : Master Thesis 
Publishing Year : 1432 AH
2011 AD 
Added Date : Tuesday, June 21, 2011 

Researchers

Researcher Name (Arabic)Researcher Name (English)Researcher TypeDr GradeEmail
شعاع جادالله الحجيليAlHijaili, Shoaa JadAllahResearcherMaster 

Files

File NameTypeDescription
 29864.pdf pdf 

Back To Researches Page