Ethiopian face Recognition System Across Ages

Face recognition is one of the most attractive research area in the field of pattern recognition and computer vision due to its interesting applications like security, access control, passport verification, human computer interaction, surveillance, etc. Face recognition systems have been proven sensitive to variations such as illumination, expression, pose and age. Facial age is complex variation in the study of face recognition that affects both shape of the face and its texture. This further affects the performance of face recognition system; it is therefore the aim of this study is to develop Ethiopian face recognition system across ages. Datasets comprising pairs of face images that were collected from different sources are used for experimentation and performance analysis of the prototype developed in this study. Accordingly, experimental result shows that face recognition system designed using Fisher Linear Discriminant analysis (FLD) demonstrates the best accuracy of 83%. However, the performance of the system is greatly affected by face variations that come due to facial age like shape, texture, facial hair, and eyeglass.