Тип публикации: доклад, тезисы доклада, статья из сборника материалов конференций
Конференция: Telegram Conference on Future Professions in the Digital Economy: Development Prospects and Social Consequence, 2020
Год издания: 2021
Идентификатор DOI: 10.1007/978-3-030-69415-9_149
Ключевые слова: asbestos, autodetection of asbestos, detecting methods, image frequency-domain analysis
Аннотация: Purpose: Asbestos is a dangerous substance, which leads to severe diseases if ingested. Therefore, it is very important to recognize asbestos and remove it from the building before work begins in it in order to protect employees in the first place and, in the second place, residents. The purpose of the research is to improve innovaПоказать полностьюtive methods of automatic detection asbestos fibers in construction materials, to speed up the process of testing samples and improve the quality of analysis, as well as to reduce the overhead and time that is wasted during testing by developers. Design/methodology/approach: In the first stage of project the authors have identified and tested a frequency-domain analysis of image for automatic detection asbestos fibers on sample images of electronic microscope. Findings: It is shown that the studied method doesn’t show an enough positive result with a high probability and can’t be used for effective testing. The results of the study showed that in this area it is reasonable to test in the next stage -the machine learning method. Outlook - further research approaches: The research is at the initial stage of development. Further research is planned using artificial intelligence to detect asbestos. © 2021, Springer Nature Switzerland AG.
Журнал: Lecture Notes in Networks and Systems
Выпуск журнала: Vol. 198
Номера страниц: 1355-1361
ISSN журнала: 23673370
Издатель: Springer Science and Business Media Deutschland GmbH