Automatic Gender Identification from Children Facial Images Using Texture Descriptors

摘要

Soft biometric such as gender significantly works to enhance the performance of biometric systems and also having applications in human-computer interaction, content-based indexing and retrieval, demographic studies, security, and surveillance. Gender identification among adults is easy as compared to in children, due to similarity of facial skin texture and appearance of faces. In this paper, we have introduced a technique to identify/classify the gender from facial images of children. In our method, we have applied three basic steps namely preprocessing, feature extraction, and classification. In preprocessing stage, face detection and normalization are performed. To extract the powerful features, we have computed different texture descriptors and after feature extraction process and S VM is applied for classification purpose. We have achieved the encouraging results in our experiments.