An efficient method for gender classification using hybrid CBR

With the advancement in hardware for face recognition, the research focus has shifted towards gender classification. In the recent years, gender classification has become an integral part of many commercial applications. It has been noted to generate monumental revenue for corporate industry among other things. Nevertheless accuracy is a case in point. In this paper, we have experimentally analyzed the accuracy and reliability of state of the art techniques for recognition and gender identification of still facial images. CBR has been used as the underlying model. The idea is to use various distance measures within, individually. Finally we have proposed a hybrid approach taking into account the pros and cons of the known experimented techniques in terms of their efficiency and accuracy.