Paper
5 January 2024 Information system for neural network analysis of x-ray Moiré images using wavelet filtering
Ihor M. Fodchuk, Mariana Borcha, Ivanna Yanchuk
Author Affiliations +
Proceedings Volume 12938, Sixteenth International Conference on Correlation Optics; 129382K (2024) https://doi.org/10.1117/12.3016096
Event: International Conference Correlation Optics (COR 2023), 2023, Chernivtsi, Ukraine
Abstract
An information system has been developed for the analysis of x-ray moiré images using artificial neural networks and wavelet filtering. Moiré images are obtained as a result of the action of a set of forces on the surface of the crystal. The analysis of the images consisted in solving the inverse problem, that is, in calculating the values of a set of forces based on the moiré image. To prevent retraining of the neural network, a control dataset of images was used. The input signals of the artificial neural network are moiré images after contour extraction, logarithmization, downscaling and wavelet filtering using Dobeshe wavelets. Due to such image processing, the training time of the neural network was reduced with a sufficiently low training error. The output signals of the neural network are the values of a set of forces that formed the moiré image. The training of an artificial neural network is performed by the method of backpropagation. The artificial neural network is implemented in the Python language in the Google Colab cloud platform. The results of testing the developed information system showed a high accuracy of calculating the values of a set of forces for simulated and experimental x-ray moiré images.
© (2024) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ihor M. Fodchuk, Mariana Borcha, and Ivanna Yanchuk "Information system for neural network analysis of x-ray Moiré images using wavelet filtering", Proc. SPIE 12938, Sixteenth International Conference on Correlation Optics, 129382K (5 January 2024); https://doi.org/10.1117/12.3016096
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Artificial neural networks

Wavelets

Crystals

X-ray imaging

X-rays

Tunable filters

Image filtering

RELATED CONTENT


Back to Top