MACHINE LEARNING BASED FACE RECOGNITION USING PARTICLE SWARM OPTIMIZATION

Abstract

Machine Learning is an interdisciplinary field that has its applications in various fields. Machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Face recognition has so many important applications that researchers have been doing a lot of work for decades to improve accuracy and speed of the face recognition algorithms. We present a systematic comparison of machine learning methods applied to the problem of Face Recognition. We report results on a series of experiments comparing recognition using minimum distance classifier, support vector machines (SVM) and minimal complexity machines (MCM). We also explored feature selection techniques by the use of Particle Swarm Optimization in addition to Principle Component Analysis. Best results were obtained by selecting features using PCA, PSO and doing classification with Minimal Complexity Machines. The system operates in real-time and obtained 96.6% accurate results on the Cohn-Kanade expression dataset.