METHODS AND MEANS OF RECOGNITION OF FLYING OBJECTS

Authors

DOI:

https://doi.org/10.31891/2307-5732-2023-325-5-150-153

Keywords:

neural networks, CNN, accuracy, concurency

Abstract

This research project revolves around the development of a sophisticated system dedicated to the recognition of objects in the sky. The analysis of the results of the application of a new image classification algorithm was carried out, which was tested on a dataset that includes images of objects in the sky. The obtained results demonstrate the high accuracy of this algorithm in image classification, and it works effectively in parallel mode. To evaluate the efficiency of the proposed algorithm, a comparative analysis of the speed of calculations with different amounts of flows was carried out. It was found that parallel computations significantly reduce the execution time of the algorithm compared to the sequential approach. One possible direction for future research could include extending the scope of the study to other types of images and using more complex neural network architectures to improve the results. The obtained results can be applied in the development of automatic image classification systems in various fields, such as medicine, security, marketing, and others.  Leveraging cutting-edge neural networks, state-of-the-art machine learning techniques, and advanced image processing technologies, this system exhibits the remarkable capability to autonomously detect and categorize a diverse array of objects within Earth's atmosphere. The dataset comprises photos of airplanes, helicopters, missiles, fighter jets, drones. One of the pivotal objectives of this endeavor is to ensure not only the efficiency but also the user-friendliness of the system. This dual focus is aimed at facilitating a seamless integration of this technology into a plethora of domains, including meteorology for more accurate weather forecasting, security for enhanced surveillance and monitoring capabilities, and scientific research within the realms of astronomy and aerospace technology, enabling groundbreaking discoveries and innovations. In conclusion, this research represents a significant leap forward in our quest to gain a deeper understanding of our ever-evolving skies and harness this knowledge for various practical applications. It underscores the transformative potential of modern technology in addressing multifaceted challenges and expanding the horizons of human knowledge.

Published

2023-10-30

How to Cite

KOSHLAN, N., KOSHLAN, N., & MEKNYKOVA, N. (2023). METHODS AND MEANS OF RECOGNITION OF FLYING OBJECTS. Herald of Khmelnytskyi National University. Technical Sciences, 325(5(1), 150-153. https://doi.org/10.31891/2307-5732-2023-325-5-150-153