MuBDA: Multimodal Biometric Data Analysis for Gender Classification Using Deep Learning Techniques

In recent years, technological advancements have led to an increased focus on security measures for individuals due to the prevalence of inappropriate behavior in society. Biometric applications play a vivacious role in the security authentication and authorization of individuals in a day-to-day life activities. These applications are used in a diversity of fields, including health care, forensic applications, and in the court of law. The proposed work emphasizes the importance of multimodal biometric traits in the demographic characteristic analysis of an individual. A total of 3510 face and offline handwritten signatures were collected from 193 male and 158 female volunteers. Transfer learning models were used to analyze the features and classify the data. The algorithms outperformed state-of-the-art techniques and produced comparable results.