High-Speed Automatic Human Face Recognition System
Sulaiman S Alhazzaa, Amani Alkhudair, Brian D Hartnett, Dexin Ren
The purpose of this project is to help develop further techniques and uses of high speed automatic facial recognition. This technology is used to detect requested people, such as criminals and missing people. Our focus is face feature extraction which is broken down into three stages. The first stage is face detection which may be performed with issues under various environments such as different sizes of the input faces, difficult lighting conditions, and multiple camera angles. To tackle these issues, we found solutions as following: for different sizes of the input faces, we recorded training images with different sizes by adjusting the distance from the recorded face to the input camera; for lighting issues, we changed the lighting of the environment that the subject was in by overexposure and underexposure the image; for camera angles, we trained the system with a large amount of images related with assorted angles of the camera. Higher angles were focused on to simulate a surveillance camera in an environment like a store or shopping center. A combination of the issues tested the outer functional limits of the program. Tests were first conducted on the team members to understand the functionality of the program. After, more individuals were added to the database. We created design criteria and wanted the speed, reliability, its security, and its legality to be the most important aspect of the program. Another main feature of the program is to extract multiple images of the human face and place the images in a database created specifically for each person. The concluding feature of the program is to compare the images of the human subject against individuals already registered in the database to quickly and accurately identify the person. The team is continuing to research the most efficient way to implement this technology.
Course Project - Undergraduate
Vijayan K Asari, Philip E Doepker, Amy T Neidhard-Doll
Primary Advisor's Department
Electrical and Computer Engineering
Stander Symposium poster
"High-Speed Automatic Human Face Recognition System" (2017). Stander Symposium Posters. 1019.