Have a personal or library account? Click to login
Design of Automated Defectoscopy Systems Cover
Open Access
|Dec 2025

Figures & Tables

Fig. 1.

An examples of metal BR (bodies of rotation)

Fig. 2.

Typical surface defects BR

Fig. 3.

Distribution function of the variability vavi, j : Vmin – minimum acceptable value of sliding dispersion variability; Vmax – maximum acceptable value; ΔVmax – uncertainty range introduced for borderline classification cases; vavi, j – difference between local maximum and minimum coefficient of variation within the sliding window

Fig. 4.

The input stream of bodies of revolution (BR) is classified into three categories: suitable (OK), defective (Defect), and undetermined (Pre-training). The undetermined group serves as an experimental base for additional supervised training. This classification allows the system to iteratively refine the decision boundary between suitable and defective BR through feedback-based adaptation
DOI: https://doi.org/10.2478/ama-2025-0069 | Journal eISSN: 2300-5319 | Journal ISSN: 1898-4088
Language: English
Page range: 611 - 616
Submitted on: Mar 5, 2025
|
Accepted on: Nov 3, 2025
|
Published on: Dec 19, 2025
In partnership with: Paradigm Publishing Services
Publication frequency: 4 issues per year

© 2025 Łukasz LEMIESZEWSKI, Janusz SZYMCZYK, Evgeny OCHIN, published by Bialystok University of Technology
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.