Fusion for Image Based Human Age Estimation

Abstract

AbstractHuman faces act as essential visual signal and reveal significant amounts of nonverbal information in human-to-human communication. In an automatic age estimation system, shape-based and texture-based features extracted from faces forms the basis of age estimation. Human age estimation with facial image analysis as an automated method houses many potential real‐world factors. This paper presents an automated age calculation framework with the help of Support Vector Regression (SVR) strategy and it highlights feature extraction with Gray Level Co-occurrence Matrix (GLCM), and Active Appearance Models (AAMs) to evaluate human age. A fused feature technique and a SVR based Cuckoo Search (CS) is proposed in order to reduce error in age estimation.