Application of Eigenvalues and Eigenvectors in Face Recognition

Abstract

Face recognition is now an essential part of human–computer interaction, surveillance systems, and biometric authentication. The eigenvalue–eigenvector based Eigenface methodology has been popular among different computing techniques because of its high performance in controlled situations and mathematical simplicity. The contribution of eigenvalues and eigenvectors to dimensionality reduction and feature extraction in face recognition is examined in this work. Principal Component Analysis (PCA) is used to convert facial images into a lower-dimensional eigenspace where robust and efficient recognition is achieved. The paper also examines developments, difficulties, and enhancements to the initial eigenface model.