Rafal GASZ, Michal TOMASZEWSKI, Slawomir ZATOR, Bogdan RUSZCZAK and Aleksandra KAWALA-STRENIUK
Opole University of Technology, Opole, Poland
Volume 2021 (8),
Article ID 37132021,
Innovations in Artificial Intelligence, Machine Learning and Intelligent Systems: 37AI 2021
Abstract
In this paper potential implementation of various digital image processing methods for the purpose of initial, semi-automated spatial object diagnostics was presented. As an example – electrical power line latticework was applied. The study involved also a thorough background study on literature related to diagnostics, which encompassed the subject matter associated with the identification of objects and detection methods of specific properties of the studied object on digital images. In this paper the authors presented their developed method of virtual photographs as a tool for support the work of technicians in the process of analysis of the photographic material collected during annual areal inspections of the lines. Additionally, this work discusses particular stages of the developed method and its software implementation. The primary objective of the virtual photograph method is to limit the number of analyzed photographs to only those, which indicate the potential risk of inconsistencies with the model – lack of an element or deformation. This paper also presents analysis of the developed method’s results for different types of photographs, when modification of the values of specific variables affected the achieved results. It also presents the directions of further research in terms of functional development of the method and potential new applications.