HUMAN FACIAL AGE ESTIMATION BY POSITIONAL TERNARY PATTERN AND GRAY-LEVEL CO-OCCURRENCE MATRIX

摘要

Human facial age estimation is done using image processing. Many applications like forensics, security, and biometrics have attracted much attention in human facial age estimation. Multiclass classification and regression problem are the existing approaches that cast facial age estimation. We propose a positional ternary pattern algorithm that inherits the craniofacial shape with wrinkle and micro texture pattern. And then Gray-Level Co-occurrence Matrix plays a major role in revealing properties of gray levels in texture image. Age estimation based on human face remains a problem in computer vision and pattern recognition. To estimate an accurate age most of the existing system is used and it requires a huge data set attached with age labels. In addition to the proposed approach we proposed the probabilistic neural network that is widely used in classification and pattern recognition problem.