Date of Award

Spring 2011

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical & Computer Engineering

Committee Director

Zia-ur Rahman

Committee Director

Jiang Li

Committee Member

Sacharia Albin

Committee Member

Yuzhong Shen

Committee Member

Dimitrie C. Popescu

Committee Member

Volkan Pehlivanoglu

Abstract

Demands for reliable person identification systems have increased significantly due to highly security risks in our daily life. Recently, person identification systems are built upon the biometrics techniques such as face recognition. Although face recognition systems have reached a certain level of maturity, their accomplishments in practical applications are restricted by some challenges, such as illumination variations. Current visual face recognition systems perform relatively well under controlled illumination conditions while thermal face recognition systems are more advantageous for detecting disguised faces or when there is no illumination control. A hybrid system utilizing both visual and thermal images for face recognition will be beneficial.

The overall goal of this research is to develop computational methods that improve image quality by fusing visual and thermal face images. First, three novel algorithms were proposed to enhance visual face images. In those techniques, specifical nonlinear image transfer functions were developed and parameters associated with the functions were determined by image statistics, making the algorithms adaptive. Second, methods were developed for registering the enhanced visual images to their corresponding thermal images. Landmarks in the images were first detected and a subset of those landmarks were selected to compute a transformation matrix for the registration. Finally, A Genetic algorithm was proposed to fuse the registered visual and thermal images. Experimental results showed that image quality can be significantly improved using the proposed framework.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/ad8f-5t84

ISBN

9781124696683

Download

Share

COinS