IDENTIFICATION OF MILITARY OBJECTS BASED ON ARTIFICIAL NEURAL NETWORKS

This paper explores modern approaches to military object recognition using neural networks (ANNs). It highlights the application of the capsule network algorithm, which improves the modeling of hierarchical relationships using advanced technologies. This paper explores the problem of identifying modern military assets using artificial neural networks (ANNs). A fundamental aspect of automated target detection is the ability to recognize objects in images acquired by reconnaissance platforms such as drones. In this context, convolutional neural networks (CNNs) play a crucial role in the analysis and classification of visual data acquired during aerial surveillance. Although CNNs are highly effective for object identification based on images, their performance is largely dependent on the availability of large training datasets. However, due to the classified nature of military infrastructure, obtaining sufficient training data remains a significant limitation. Therefore, insufficient training data can significantly reduce the performance of the NNN. To solve this problem, a multi-layer CapsNet platform was chosen, specifically designed for military object recognition with a small trai