Сучасні інформаційні системи

Постійне посилання на розділhttps://repository.kpi.kharkov.ua/handle/KhPI-Press/62915

У журналі публікуються результати досліджень з експлуатації та розробки сучасних інформаційних систем у різних проблемних галузях.

Рік заснування: 2017. Періодичність: 4 рази на рік. ISSN 2522-9052 (Print)

Новини

Включений до "Переліку наукових фахових видань України, в яких можуть публікуватися результати дисертаційних робіт на здобуття наукових ступенів доктора і кандидата наук" (технічні науки) наказом Міністерства освіти і науки України від 04.04.2018 № 326 (додаток 9, п. 56).

Переглянути

Результати пошуку

Зараз показуємо 1 - 4 з 4

The issue of training of the neural network for drone detection
(Національний технічний університет "Харківський політехнічний інститут", 2024) Hashimov, Elshan; Khaligov, Giblali
In the article, the issue of mistaking birds for UAVs, which is considered one of the main problems in the process of detecting unmanned aerial vehicles (UAVs) through cameras, was considered. The object of the study is a method of training an artificial intelligence model so that the detection system does not mistake drones for birds. In order to solve this problem, as an alternative to creating a new complex model, the retraining method was used by making some changes to the previously developed model. The aim of the study is to train an artificial intelligence model using certain methods to detect drones and prevent them from being mistaken for birds.
Intrusion detection model based on improved transformer
(Національний технічний університет "Харківський політехнічний інститут", 2024) Gavrylenko, Svitlana; Poltoratskyi, Vadym; Nechyporenko, Alina
The object of the study is the process of identifying the state of a computer network. The subject of the study are the methods of identifying the state of computer networks. The purpose of the paper is to improve the efficacy of intrusion detection in computer networks by developing a method based on transformer models. The results obtained. The work analyzes traditional machine learning algorithms, deep learning methods and considers the advantages of using transformer models. A method for detecting intrusions in computer networks is proposed. This method differs from known approaches by utilizing the Vision Transformer for Small-size Datasets (ViTSD) deep learning algorithm. The method incorporates procedures to reduce the correlation of input data and transform data into a specific format required for model operations. The developed methods are implemented using Python and the GOOGLE COLAB cloud service with Jupyter Notebook. Conclusions. Experiments confirmed the efficiency of the proposed method. The use of the developed method based on the ViTSD algorithm and the data preprocessing procedure increases the model's accuracy to 98.7%. This makes it possible to recommend it for practical use, in order to improve the accuracy of identifying the state of a computer system.
Development of control laws of unmanned aerial vehicles for performing group flight at the straight-line horizontal flight stage
(Національний технічний університет "Харківський політехнічний інститут", 2023) Barabash, Oleg; Kyrianov, Artemii
The article proposes an improved approach to controlling groups of unmanned aerial vehicles (UAVs) aimed at increasing the overall efficiency and flexibility of the control process. The use of a heterogeneous external field, which varies both in magnitude and direction, allows achieving greater adaptability and accuracy in controlling a group of UAVs. A vector field for unmanned aerial vehicles determines the direction and intensity of the vehicles' movement in space. Such vector fields can be used to develop UAV control laws, including determining optimal flight paths, controlling speed, avoiding obstacles, and ensuring coordination of a group of UAVs. The subject of the study is the methods of controlling groups of autonomous UAVs, where each vehicle may have different speeds and flight directions. To solve this problem, various methods of using a heterogeneous field have been developed and proposed. Instead of using a homogeneous field that provides a constant flight speed, a vector field is used that adapts to different conditions and characteristics of the vehicles in the group. This method allows for effective group management, ensuring the necessary coordination and interaction between the vehicles. An analysis of recent research and publications in the field of autonomous system control indicates the feasibility of using machine learning, vector fields, and a large amount of data to successfully coordinate the movement of autonomous systems. These approaches make it possible to create efficient and reliable control systems. The aim of the study is to develop laws for controlling the movement of a group of autonomous unmanned aerial vehicles at the stage of straight-line horizontal flight based on natural analogues to improve the efficiency and reliability of their coordinated movement in different conditions. The main conclusions of the research are that the proposed method of controlling groups of UAVs based on a heterogeneous field can be implemented. It takes into account a variety of vehicle characteristics and environmental conditions that are typical for real-world use scenarios. This work opens up prospects for further improving the management of UAV groups and their use in various fields of activity. The article emphasises the relevance of technology development for autonomous unmanned systems, especially in the context of autonomous transport systems.
Efficiency of supplementary outputs in siamese neural networks
(Національний технічний університет "Харківський політехнічний інститут", 2023) Melnychenko, Artem; Zdor, Kostyantyn
In the world of image analysis, effectively handling large image datasets is a complex challenge that requires using deep neural networks. Siamese neural networks, known for their twin-like structure, offer an effective solution to image comparison tasks, especially when data volume is limited. This research explores the possibility of enhancing these models by adding supplementary outputs that improve classification and help find specific data features. The article shows the results of two experiments using the Fashion MNIST and PlantVillage datasets, incorporating additional classification, regression, and combined output strategies with various weight loss configurations. The results from the experiments show that for simpler datasets, the introduction of supplementary outputs leads to a decrease in model accuracy. Conversely, for more complex datasets, optimal accuracy was achieved through the simultaneous integration of regression and classification supplementary outputs. It should be noted that the observed increase in accuracy is relatively marginal and does not guarantee a substantial impact on the overall accuracy of the model.