Wrinkles energy based age estimation using discrete cosine transform

Abstract

Human's faces are true representative of a human and information like gender, age and origin about a human. Age estimation on the bases of facial image is a complex phenomenon for the computer vision systems to predict human's age effectively. Facial image based Age estimation analysis has a lot of real world applications e.g. human computer interaction (HCI) and multimedia communication. In this paper a new method to predict human's age is presented. The proposed method is focusing on cheek's area and wrinkles analysis using discrete cosine transform (DCT). As the wrinkles on human face increase with increasing age, so it has been used as discriminating factor for age estimation. DCT is applied on the wrinkle detected. The energy in higher coefficients of DCT increases with the increasing wrinkle in other words with increasing age.