Face Image Based Gender Classification Using Min-max Modular Classifier

The biological characteristics are the inherent attributes of human beings, which have strong self-stability and individual independency. Therefore, it has been a common research problem to apply the individual biological characteristics into machine intelligence and make the computer more intelligent as human beings. Recently, several different biological characteristics have been applied into pattern recognition including face recognition, fingerprint recognition, iris recognition, palm print recognition and etc. As a representative, face recognition has become very popular and been improved a lot. Gender estimation is a subset problem of face recognition which can be used for not only business statistical applications and robotology but also for evaluation about performance of large-scale classifiers. Similar as face recognition, gender estimation consists of three main parts - image preprocessing, feature extraction and pattern classification. In this paper, we mainly concentrated on the feature extraction and pattern classification parts and especially analyzed the performances of applying Min-max Modular support vector machine into the gender estimation.Image preprocessing mainly focuses on image geometry normalization, intensity histogram equalization and face area masking. These will complete the image normalization and help improve the image quality, decrease the computation complexity and therefore enhance the recognition accuracy and accelerate the convergence speed.In the image feature extraction stage, both global image features and local image features are introduced in this paper including gray-scale pixel values, Gabor Wavelet Filter, Local Binary Pattern, Scale Invariant Feature Transform and etc. In our experiments we focused more on Scale Invariant Feature Transform method and analyzed the detailed performances of these different methods for representing gender information.In the process of gender estimation, our main efforts are about using Min-max Modular Classifier to discriminate gender classes. Min-max Modular classifier is composed of two parts: task decomposition and sub-problem combination. This Min-max Modular scheme is based on a 'divide-and-conquer' strategy and it can divide a large-scale problem into several...