Szczegóły publikacji
Opis bibliograficzny
Aligned marker space for vision-based detection of damage in structures with one fixed end / Ziemowit DWORAKOWSKI, Paweł ZDZIEBKO, Kajetan DZIEDZIECH, Krzysztof HOLAK // W: European Workshop on Structural Health Monitoring : EWSHM 2022 : [10th European Workshop on Structural Health Monitoring : Palermo, Italy, 4-7 July 2022], Vol. 2 / eds. Piervincenzo Rizzo, Alberto Milazzo. — Cham : Springer Nature Switzerland, cop. 2023. — (Lecture Notes in Civil Engineering ; ISSN 2366-2557 ; vol. 254). — ISBN: 978-3-031-07257-4; e-ISBN: 978-3-031-07258-1. — S. 446–453. — Bibliogr., Abstr. — Publikacja dostępna online od: 2022-06-16
Autorzy (4)
Słowa kluczowe
Dane bibliometryczne
ID BaDAP | 141381 |
---|---|
Data dodania do BaDAP | 2022-08-30 |
DOI | 10.1007/978-3-031-07258-1_46 |
Rok publikacji | 2023 |
Typ publikacji | materiały konferencyjne (aut.) |
Otwarty dostęp | |
Wydawca | Springer |
Czasopismo/seria | Lecture Notes in Civil Engineering |
Abstract
There are many methods for beam-like structures’ vision-based damage detection, usually based on deflection shape analysis. However, most engineering structures have geometries for which such a straightforward damage detection approach is not feasible. This paper introduces the Aligned Marker Space (AMS) method to close this gap and allow using straightforward deflection-difference-based damage detection methods for various civil engineering structures. The method involves transforming vision data into a new morphed space to straighten complex geometries along one dimension. Data transformed using this approach can be used as an input to various damage detection algorithms, including local (e.g., second-derivative-based or wavelet-based algorithms) and global approaches (e.g., trend detection and line-segments method). The method is verified based on data acquired in numerical simulations and in a set of practical experiments. A Blender-based approach is used to build plausible simulated data with a high resemblance to real-life examples. These examples are then validated using complex geometries in a series of physical experiments under laboratory conditions.