Advancements in Facial Image Processing for Human Age and Gender Identification

With the increased usage of social media and associated cyber frauds, current scenario demands for authenticated information. This in turn necessitates for the valid and verified information of individuals from various resources which are available online. Various examples of cyber fraud include: fake banking transactions, unauthorized access to various organizations’ resources, fake online proxy for absentees through biometric system etc. To counter such issues, in this paper we have targeted facial image processing for Precise age and correct gender recognition which are the most crucial factors in the digital era. Thus, in this study, we have constructed two distinct approaches to figure out human age and gender from facial with a satisfactory runtime and efficiency. Our method uses image processing to examine specific features from facial images of people in different age groups that have been adjusted. It then creates binary masks, performs edge detection, and measures wrinkle densities. To make assessing easier, we developed the ADIENCE facial database, containing images of males and females spanning various age ranges. The database holds twenty-six thousand plus photos featuring more than two thousand individuals sorted by gender and grouped into eight age categories. This special database has been used to train computer program for identification of individual's face based on their age and gender. This has been done by using Convolutional Neural Network (CNN) which makes it easier for program to recognize age and gender from pictures of people's faces. This is definitely going to help in various domains where security is top concern. This can also be helpful in healthcare domain for medical diagnoses.