Facial Image Analysis for Identification of Gender: A Machine Learning based Approach

Abstract

Facial biometrics for gender classification is known to significantly improve the efficiency of person identification in biometric access control models. It not only increases the speed but also improves the accuracy. It reduces the effort of finding a match for a face in the databases to almost a half. The process like any pattern recognition framework needs to extract the useful features. A novel gender identification algorithm is proposed which is based on the facial features of a person. Viola Jones object detection technique is used for face detection from an image, and the relevant facial features are extracted using Topographic Independent Component Analysis. An SVM based classifier is trained using the calculated feature vector to determine whether the person in the image is a male or a female.