Discipline: Computer Sciences and Information Management
Subcategory: Computer Science & Information Systems
Cody Butler - Norfolk State University
With criminal activities such as the Boston Marathon Bombing and the Charleston Church Shooting, the Federal Bureau of Investigation (FBI) is attempting to combat identification issues with facial recognition technology. Facial recognition software enables us to identify or verify a face in a digital image and is currently used in most cameras of malls, banks, airports, and homes. Although many advances have been made with facial recognition technology, better accuracy is still needed in the analysis done by facial recognition algorithms. Challenges occur with the images including pose of the photo, presence or absence of structural components such as eyeglasses, obstruction of the face, facial expression, orientation or additional faces in the photograph. Although researchers have worked to resolve some of these problems, most solutions involve expensive commercial software and have limited accuracy. Furthermore, most biometric software is not designed for non-technical users. This research involves development of a graphical user interface (GUI) and design of an alternative facial recognition algorithm that improves usability and accuracy of the Open Source Biometric Recognition software (OpenBR). A GUI and biometric algorithm was implemented using Qt Creator and the C++ language. Preliminary preparation of images was done using 3D modeling software and compared to an existing stored image resulting in textual indication of a match or nonmatch. Enhancements to the facial recognition software include allowing the user to select only one photograph and then have the other one appear of the person that it matches along with the textual information that currently displays. The future work for this project is to be able to integrate the facial recognition software with 3D graphics into a single software application rather than two separate processes.
References: Wagner, A., Wright, J., Ganesh, A., Zhou, Z., Mobahi, H., Ma, Yi. ‘Toward a Practice Face Recognition System: Robust Alignment and Illumination by Sparse Representation.’ IEEE Transaction on Pattern Analysis and Machine Intelligence, vol. 34, no.2 February 2012.Web. June 5, 2015.
Funder Acknowledgement(s): Establishing a Massie Chair of Excellence in Cyber Security at Norfolk State University, Department of Energy Grant # F1040025 for financially supporting the research.
Faculty Advisor: Felicia Doswell,