The thesis presents a convolutional neural network (CNN) model for the automatic classification of military vehicles based on images. The study is based on a dataset of over 14,000 images, divided into training and testing sets, covering 10 categories including tanks, artillery, infantry fighting vehicles, and anti-aircraft systems. The results confirm the effectiveness of deep learning in recognizing military objects. The proposed approach has potential applications in military analytics, drone-based surveillance systems, automated target detection, and defense-related research.
Somriakov et al. (Mon,) studied this question.