Face recognition and age determination

• Main Author: Ma, Simin
• Other Authors: Er Meng Joo
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/64305
• Institution: Nanyang Technological University

The project aims to combine face recognition and age determination to assist healthcare robots for their services. It focuses on the combination of real-time face recognition system and age determination system in order to recognize patients and determine their ages to facilitate further actions. A brief introduction to the background, objective, and scope is done in the first chapter. In the second chapter, literature review about various face recognition and age determination techniques are discussed. In the proposed methodology, Matrix Laboratory (MATLAB) is used to simulate a robot’s controller, and Webcam is used to simulate a robot’s vision system. Face recognition involves two steps, which are face detection and face recognition. Face detection utilizes Viola-Jones algorithm and face recognition utilizes the mean shift algorithm and Principle Component Analysis (PCA) method. A fast and robust age determination technique is integrated into the system to estimate ages by examining local facial features. The overall results of the proposed system meet the requirements of a proper face recognition and age determination system, with a 95.7% accurate rate for face recognition and an average 80.7% accuracy for age determination. Some limitations and recommendations are discussed in the last chapter.