Gabor based gender classification with classifier independent feature selection

Abstract

The study presents an efficient gender classification technique. The gender of a facial image is the most prominent feature, and improvement in the existing gender classification methods will result in the high performance of the face retrieval and classification methods for large repositories. The method presented in this paper selects the effective set of Gabor features which are ranked on the base of entropy and are merged with mean Gabor feature values. This forms an effective feature vector to be used by the classifiers. The method presented in this paper enjoys the features like high speed and low space requirements. Proposed technique has been tested on different datasets. It shows good performance as comparative to existing techniques. © 2011 Springer-Verlag.