Journal of Radio Electronics. eISSN 1684-1719. 2025. ¹11
Full text in Russian (pdf)
DOI: https://doi.org/10.30898/1684-1719.2025.11.39
Evaluation of the efficiency of methods
for reconstructing distorted images
at different levels of quantization
A.V. Kokoshkin
Kotelnikov Institute of Radioengineering and Electronics of RAS, Fryazino Branch,
141190, Russia, Fryazino, pl. Vvedenskogo, 1
The paper was received September 19, 2025.
Abstract. The article presents an assessment of the efficiency of reconstruction methods based on the adaptation of the well-known metric “mean absolute percentage error” adapted to the tasks of digital image processing. The qualitative similarity of the estimates for the metrics “Adaptive Mean Absolute Percentage Error” and “Structural Similarity Index Measure” is shown, and their quantitative differences suggest their complementarity. The results of the comparative analysis indicate the high quality of the restoration methods used: the Adaptive Reference Image Method (ARIM), the Constrained Renormalization Method, and the Modification of the Wiener Filter based on ARIM.
Key words: image processing, quality assessment metrics, quantization levels, image reconstruction methods.
Financing: The work was carried out within the framework of the state task of the Kotelnikov Institute of Radioengineering and Electronics (IRE) of Russian Academy of Sciences.
Corresponding author: Kokoshkin Alexander Vladimirovich, shvarts65@mail.ru
References
1. Gonzalez R.C., Woods R.E., Digital Image Processing. – NJ.: Prentice Hall, International Version 3rd Edition. – 2012. – 1071 p.
2. Bates R.T., McDonnell M.J. Image restoration and reconstruction. – Oxford University Press, Inc. – 1986. – 288 p.
3. Pratt W.K. Digital image processing. – Wiley-interscience, – 2007. – 738 p.
4. Gruzman I.S. et al. Digital image processing in information systems – Novosibirsk: NSTU. – 2002. – 352 p. (In Russian)
5. Hanke M., Neubauer A., Scherzer O. A convergence analysis of the Landweber iteration for nonlinear ill-posed problems // Numerische Mathematik. – 1995. – V. 72. – ¹. 1. – P. 21-37. https://doi.org/10.1007/s002110050158
6. LeCun Y., Bengio Y., Hinton G. Deep learning // nature. – 2015. – V. 521. – ¹. 7553. – P. 436-444. https://doi.org/10.1038/nature14539
7. Gulyaev Yu.V. et al. Correction of the spacial spectrum distorted by the optical system using the reference image method. Part 1 – 3. // Journal of Radio Electronics. – 2013. – ¹.12. (In Russian)
8. Kokoshkin A.V. et al. Using the method of renormalization limited to restore the distorted image in the presence of interference and noise with unknown parameters. // Journal of Radio Electronics. – 2015. – ¹.7. (In Russian)
9. Kokoshkin A.V. Modification of the Wiener filter based on the reference image method. // Journal of Radio Electronics. – 2024. – ¹11. https://doi.org/10.30898/1684-1719.2024.11.10 (In Russian)
10. Kokoshkin A.V. et al. Comparison of objective methods of assessing quality of digital images. // Journal of Radio Electronics. – 2015. – ¹. 6. (In Russian)
11. Kokoshkin A.V. Estimation of spectral similarity of digital images. // Journal of Radio Electronics. – 2020. – ¹. 8. https://doi.org/10.30898/1684-1719.2020.8.4 (In Russian)
12. Kokoshkin A.V. Sparse Image Interpolation Methods Working in the Frequency Domain // Journal of Radio Electronics. – 2023. – ¹. 3. https://doi.org/10.30898/1684-1719.2023.3.10 (In Russian)
13. Kokoshkin A.V., Novichikhin E.P. Evaluation of images quality obtained by remote sensing. // RENSIT: Radioelectronics. Nanosystems. Information Technology – 2023. – V.15. – ¹. 3. – P.327-334. https://doi.org/10.17725/rensit.2023.15.327
For citation:
Kokoshkin A.V. Evaluation of the efficiency of methods for reconstructing distorted images at different levels of quantization // Journal of Radio Electronics. – 2025. – ¹. 11. https://doi.org/10.30898/1684-1719.2025.11.39 (In Russian)