Intelligent system of recognition of the type and geolocation of military equipment using machine learning technologies

DOI: 10.31673/2409-7292.2024.020004

Abstract

The development of intelligent systems for military needs is a key factor in the superiority of the technical equipment of the defense forces of Ukraine over the enemy. Modern technical support should be quick to produce results, accurate, portable, compact and work offline. Therefore, the development of intelligent systems capable of recognizing images and effectively locating military equipment will increase the accuracy and speed of intelligence operations, reducing intelligence risks and increasing the overall level of security. The task of object detection is to automatically identify and locate the object to be detected in an image or video. Traditional object detection methods mainly use hand-crafted features to train classifiers. However, in recent years, with the development of convolutional neural networks, object detection methods based on deep learning have gradually attracted attention. In this study, an analysis and comparative study of the effectiveness of the use of machine learning technologies and convolutional neural networks for the identification of objects of military equipment based on photos and videos was carried out. The focus of the study is on ensuring key parameters such as detection accuracy, query processing speed and overall system reliability. An intelligent system based on machine learning technologies using CoreML technology and the Swift programming language has been developed, which is designed to determine the type and geolocation of military equipment. This system is integrated into the mobile connection and can be used without internet connection. The research is aimed at solving practical problems in increasing the accuracy of enemy military equipment detection systems.

Keywords: intelligent system, machine learning, object identification, recognition accuracy.