Моделювання нейромережевого алгоритму для аналізу дефектів відеопотоку
Loading...
Date
2025
Journal Title
Journal ISSN
Volume Title
Publisher
Український державний університет науки і технологій, ННІ ≪Інститут промислових та бізнес технологій≫, ІВК ≪Системні технології≫, Дніпро
Abstract
UKR: В роботі зроблено огляд на існуючі наукові роботи з аналізу відео на виявлення дефектів та технологій цифрового мовлення. Розглядається створення дата-сета, розподілу даних по вибіркам. Описується навчання нейромережевих моделей та отримання результатів навчання. Реалізації методу використання навченої нейронної мережевої моделі.
ENG: The paper provides an overview of existing scientific works on video analysis for defect detection and digital broadcasting technologies. The creation of a data set, data distribution by samples are considered. The training of neural network models and obtaining training results are described. Implementation of the method of using a trained neural network model. From the results obtained, we can draw an unfortunate conclusion that the trained model YOLOv8x.pt has the highest accuracy in detecting defects. After obtaining the best results of model training, it was found that the highest resolution affects the quality of error detection only up to a certain value. On this training sample, a sufficient resolution for image detection is 540x540 pixels. The computing resources of the Google Colaboratory service are limited for free use, so the number of training epochs was reduced for the YOLOv8x.pt and yolov5x6u.pt models. Even with fewer training epochs, the YOLOv8x.pt model was able to show the best result. For further use, the trained neural network model YOLOv8x.pt will be used.
ENG: The paper provides an overview of existing scientific works on video analysis for defect detection and digital broadcasting technologies. The creation of a data set, data distribution by samples are considered. The training of neural network models and obtaining training results are described. Implementation of the method of using a trained neural network model. From the results obtained, we can draw an unfortunate conclusion that the trained model YOLOv8x.pt has the highest accuracy in detecting defects. After obtaining the best results of model training, it was found that the highest resolution affects the quality of error detection only up to a certain value. On this training sample, a sufficient resolution for image detection is 540x540 pixels. The computing resources of the Google Colaboratory service are limited for free use, so the number of training epochs was reduced for the YOLOv8x.pt and yolov5x6u.pt models. Even with fewer training epochs, the YOLOv8x.pt model was able to show the best result. For further use, the trained neural network model YOLOv8x.pt will be used.
Description
К. Островська: ORCID 0000-0002-9375-4121
Keywords
нейромережа, згорткова нейронна мережа, дата-сет, дефект, відеофайл, відеотрансляція, відеомовлення, neural network, convolutional neural network, data set, defect, video file, video broadcast, video broadcast, КІТС
Citation
Островська К. Ю., Зимогляд А. Ю., Козакова А. О. Моделювання нейромережевого алгоритму для аналізу дефектів відеопотоку. Системні технології. Дніпро, 2025. Т. 3, № 158. С. 110–116. DOI: https://doi.org/10.34185/1562-9945-3-158-2025-11.