Combination of Local Binary Pattern and Face Geometric Features for Gender Classification from Face Images

In the recent time bioinformatics take wide field in image processing and computer vision. Gender classification is essentially the task of identifying the person gender based on the facial image. Currently the gender classification by facial images becomes very popular due to the current visual instruments. There are different algorithms of gender classification, and each algorithm has a different approach to extract the facial feature from the input image and perform the classification. However, the single type face feature cannot be enough to represent the detailed in facial images. In this paper, we propose a new approach which consists in combining the local binary patterns (LBP) and the face geometric features to classify gender from the face images. The Histogram equalization is used to adjust the contrast of the input image. For encoding the gray level pixel, the LBP is used as a binary quantization, then the face GLCMs are used to extract the geometric structure of the face image. For gender classification, the Support Vector Machine is used as the classifier. The face images from AT&T face dataset is used to perform the experiments. The experimental results show that the application of both LBP, and the GLCMs features improves the performance the classification of gender in face images.