Поиск по всему репозиторию:
Сборник материалов конференции по статьям: Недавние поступления
Отображаемые элементы 21-40 из 48
-
Возможности измерения плотности тепловых потоков
(БрГТУ, 2024)In modern conditions, measurements of heat flux density are becoming important in engineering. They are necessary in thermophysical experiments devoted to the study of the properties of substances and heat exchange processes, as well as for diagnostics of industrial heat and power equipment and control ...2025-01-08
-
Применение нейросетей в повседневной жизни: распознавание и детекция лиц
(БрГТУ, 2024)This article describes the mechanism of neural networks, and in particular the mechanism of facial recognition. The basic principle of face recognition with such stages as detection, normalization, extraction is highlighted.2025-01-08
-
Телеграм-бот с искусственным интеллектом «Электронный консультант БрГТУ»
(БрГТУ, 2024)Telegram-bots are an integral part of modern digital services, and their role in facilitating communication, automating tasks, and providing real-time responses is increasing. The use of artificial intelligence (AI) in such bots ensures their ability to handle complex tasks and provide personalized ...2025-01-08
-
Оптимизация рекламных стратегий с использованием рекуррентной нейронной сети с долгой краткосрочной памятью
(БрГТУ, 2024)Modern e-commerce advertising requires the use of advanced data analysis technologies to improve the effectiveness of user interaction. One of the effective approaches to managing advertising companies is the use of neural networks such as Long Short-Term Memory (LSTM). In this paper, we consider the ...2025-01-08
-
Одноплатные компьютеры на базе процессорной архитектуры RISC-V в системах компьютерного зрения
(БрГТУ, 2024)This paper proposes the use of the Milk-V Duo series SBC for building computer vision subsystems with a comparative performance characteristic of each device to solve the problem of detecting objects in an image using the Yolo v5 neural network.2025-01-08
-
Методика построения карты плотности распределения ядрышковых структур на основе данных цифровой микроскопии
(БрГТУ, 2024)This paper deals with the technique of constructing a density map of distribution of nucleus structures in cells based on digital microscopy data using modern deep learning methods. Nucleus organisers, which are key chromosome sites for rRNA synthesis, play an important role in studying the proliferative ...2025-01-08
-
Стационарный теплообмен между шарами и потоком воздуха
(БрГТУ, 2024)This scientific article is devoted to the experimental study of heat exchange between balls and air flow under stationary conditions. The main objectives of this study were to clarify the available data on stationary heat exchange.2025-01-08
-
Подход к оптимизации выкладки товаров с использованием искусственного интеллекта
(БрГТУ, 2024)An innovative approach to optimizing product placement in retail using artificial intelligence is proposed. The method is based on the application of the LLaMA 3 family Large Language Model for analyzing and interpreting regulatory documents, as well as optimization algorithms for creating effective ...2025-01-08
-
Применение интеллектуальных технологий для оценки эффективности использования матричных двумерных кодов при автоматизации управления основными средствами
(БрГТУ, 2024)The study identified the types of defects that have the greatest impact on the accuracy and efficiency of fixed assets when using different methods. Certain parameters made it possible to establish the impact of errors and the costs of correcting them, which confirmed the economic feasibility of using ...2025-01-08
-
Оценка точности моделирования тепловых процессов при увеличении количества конечных элементов в исследуемой области
(БрГТУ, 2024)The finite element method (FEM) is a vital tool in the engineering field, enabling detailed temperature distribution analysis in complex systems like processor radiators. This study highlights the process and methodology of FEM, emphasizing its flexibility and precision in modeling intricate geometries ...2025-01-08
-
Анализ переноса тепловой энергии в гидродинамической среде
(БрГТУ, 2024)This article deals with the modeling of thermal energy transfer processes in hydrodynamic environments. This research is important for the development of efficient cooling systems, especially in the context of radio electronics and mobile devices. The main heat transfer mechanisms are analyzed, as ...2025-01-08
-
Применение метода матриц переноса для расчетов гетероструктур InGaAs/AlGaAs/GaAs
(БрГТУ, 2024)The method of transfer matrices for calculating optical properties of InGaAs/AlGaAs/GaAs heterostructures with different number of quantum wells is considered. The maximum optical confinement factor depending on the number of quantum wells and thicknesses of waveguide layers are calculated.2025-01-08
-
Моделирование исходно открытого гетеропереходного полевого транзистора на основе нитрида галлия
(БрГТУ, 2024)AlGaN ternary alloys with a band gap of 3.4 to 6.2 eV are promising materials for the construction of various electronic devices: diodes, transistors, lasers, microwave circuits. Replacing silicon with GaN allows several times to increase the operating temperature, the cutoff frequency, and to reduce ...2025-01-08
-
Анализ классических и нейросетевых методов обнаружения дыма на видеоизображениях
(БрГТУ, 2024)The paper reviews existing methods for detecting smoke in video images, taking into account accuracy and performance. Smoke images and their sequences have visual features: specific brightness and color characteristics, indefinite shape, directed low-speed movement, reduction in the energy of ...2025-01-08
-
Экспериментальный выбор гиперпараметров сверточной нейронной сети DenseNet-121 для реидентификации людей по изображениям
(БрГТУ, 2024)Hyperparameter choice is one of the key methods to improve accuracy for people re-identification. For this task, we used a two-stage approach. Our methodology allowed us to find among many possible values the most effective combination for batch size, learning rate and number of epochs of CNNs training ...2025-01-08
-
Линейное сглаживание эмпирической функции вероятности для малых выборок данных
(БрГТУ, 2024)The main advantage of nonparametric statistics, including ordinal (rank) statistics, is their independence both from the parameters and from the type of distribution of the random variable. The proposed two methods for estimating the empirical distribution function, based on order statistics: a reliable ...2025-01-08
-
Моделирование и визуализация данных учебных планов образовательных программ
(БрГТУ, 2024)The paper discusses an approach to analyzing the curricula of educational programs of higher education in the context of current federal state educational standards. Based on the competency-based approach, a model for representing the curriculum in the form of a mixed graph is proposed. This approach ...2025-01-08
-
Методы машинного обучения для распознавания потенциальных антропогенных угроз в системах «умного» города
(БрГТУ, 2024)In this article the description of the problem of recognizing potential anthropogenic threats is given. The methods used to recognize potential anthropogenic threats are analyzed.2025-01-08
-
Разработка методики сбора и анализа телеметрических данных грузового автомобиля
(БрГТУ, 2024)A method for collecting and analyzing data from standard and additional sensors, as well as the truck CAN bus, was developed, on the basis of which a method for intelligent data analysis was created and tested at an operating transport company. The result of the method is an analytical assessment of ...2025-01-08
-
Перспективы развития бортовых встроенных систем сельскохозяйственной техники
(БрГТУ, 2024)The onboard embedded systems of agricultural machinery manufactured by JSC Tsvetotron are considered: an information management system, a multifunctional terminal, an on-board information management system, a preservative application system, a universal mass control system for the components of the ...2025-01-08