Modified CNN Model for Classifying Gender of Thermal Images Using Cloud Computing

Abstract

The utilization of thermal images has become widespread in various applications, particularly for thermal examination and night surveillance systems. Although many details associated with thermal imaging are virtually obscured or unclear, thermal imaging offers numerous benefits, including the ability to determine an individual's gender even in cases where human vision is impaired. Based on thermal faces, a convolutional neural network (CNN) has been developed to distinguish between people's genders. Since cloud computing offers a suitable setting and gained attraction, its concept has been adopted. Thermal databases were used for the experiments. The suggested model provides an overall accuracy of 99% and a gender categorization precision of 97.75%.