An Unsupervised Technique for Automatic Face Detection and Extraction

Automatic face detection is the essential part of the facial expression recognition (FER) systems. Before investigating the facial expressions, it is compulsory to detect and extract the faces first from the expression frames. Existing methods often involve modeling of the face detection that normally necessitates huge amount of training data and cannot efficiently tackle changes over time. In this paper, an unsupervised technique based on active contour (AC) model is adopted in order to detect and extract the human faces automatically from the expression frames. In this model, the combination of two energy functions like Chan-Vese (CV) energy and Bhattacharyya distance functions were exploited that not only minimize the dissimilarities within the object (face) but also maximize the distance between the object (face) and background. The developed method is more robust to noise and illumination that are typical issues in FER systems. The proposed AC model is an unsupervised technique; means no training data is required. The developed approach achieved best results than of conventional CV AC model.