PSO-GA Based Optimized Feature Selection Using Facial and Clothing Information for Gender Classification

Gender classification is a fundamental face analysis task. In previous studies, the focus of most researchers has beenon face images acquired under controlled conditions. Real-world face images contain different illumination effects andvariations in facial expressions and poses, all together make gender classification a more challenging task. In thispaper, we propose an efficient gender classification technique for real-world face images (Labeled faces in the Wild).In this work, we extracted facial local features using local binary pattern (LBP) and then, we fuse these features withclothing features, which enhance the classification accuracy rate remarkably. In the following step, particle swarmoptimization (PSO) and genetic algorithms (GA) are combined to select the most important features' set which moreclearly represent the gender and thus, the data size dimension is reduced. Optimized features are then passed tosupport vector machine (SVM) and thus, classification accuracy rate of 98.3% is obtained. Experiments are performedon real-world face image database.