Real-time face recognition using computational intelligence techniques

• Main Author: Tee, Whye Sheng
• Other Authors: Er Meng Joo
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Home entertainment systems; DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Biometrics
• Online Access: http://hdl.handle.net/10356/64078
• Institution: Nanyang Technological University

Going into 21st century, technologies related to computers will become more connected and attached to our lives. Their functions range from computation and calculation works up to the latest technology of biometrics, artificial intelligence and face recognition. Among these technology, face recognition is very important as it can be used to recognize dangerous person in CCTV, recognize an employee that is accessing certain restricted area protected by security system and etc. The aim of this project is to develop a real-time face recognition system that can detect and recognize multiple faces on the screen. As images are essentially large in their sizes, it requires heavy computation work to do some image processing techniques on the images directly. This will resulted in a very long computation time that slows down the face recognition system. In order to develop a real-time face recognition system, Principal Component Analysis (PCA) and Fisher’s Linear Discriminant Analysis (FLD) are first applied on the image data to reduce the size and hence the computation time needed. The images with smaller size are used to train a Radial Basis Function (RBF) Neural Network and the network will be used to recognize different people. Using these techniques, a real-time face recognition system is developed and it is able to recognize people with upright and frontal faces very well. However, some problems were encountered when the faces were rotated and resulting in some misidentifications. The developed real-time face recognition system can be applied to a lot of applications such as attendance taking in the classes, detection of human faces at bank or immigration counters, tagging people’s photos in entertainment program, and etc.