Analysis of the results of a computational experiment to restore the discontinuous functions of two variables using projections. II / Lytvyn, / Lytvyn. (2022)
Ukrainian

English  Cybernetics and Systems Analysis   /     Issue (2022, 58 (1))

Lytvyn O.M., Lytvyn O.G.
Analysis of the results of a computational experiment to restore the discontinuous functions of two variables using projections. II

This article continues a series of publications under the same name. It performs further improvement of the method for restoring discontinuous functions of two variables using projections to improve the accuracy of approximation without the Gibbs phenomenon for the case, where the discontinuity lines are the boundaries of squares nested into each other. We consider the case of the discontinuity lines having angular points, where the derivative along the normal is undefined. A discontinuous spline is constructed so that the difference between the function being approximated and this spline is a differentiable function. This function is restored using finite Fourier sums whose Fourier coefficients are found using projections. Analysis of the computing results confirmed the theoretical statement of the study. © 2022, Springer Science+Business Media, LLC, part of Springer Nature.

Keywords: computed tomography, differentiability class, discontinuous function, discontinuous spline, Fourier sum, Gibbs phenomenon, Computerized tomography, Fourier transforms, Interpolation, Restoration, Computed tomography, Computing Experiments, Differentiability, Differentiability class, Discontinuity lines, Discontinuous functions, Discontinuous spline, Fourier, Fourier sum, Gibbs phenomena, Fourier analysis


Cite:
Lytvyn O.M., Lytvyn O.G. (2022). Analysis of the results of a computational experiment to restore the discontinuous functions of two variables using projections. II. Cybernetics and Systems Analysis, 58 (1), 110–121 . doi: https://doi.org/10.1007/s10559-022-00469-8 http://jnas.nbuv.gov.ua/article/UJRN-0001301511 [In Ukrainian].


 

Institute of Information Technologies of VNLU


+38 (044) 525-36-24
Ukraine, 03039, Kyiv, Holosiivskyi Ave, 3, room 209