Izvestiya Vysshikh Uchebnykh Zavedenii. Elektronika home

Publication of the journal

The section is currently being updated

The implementation of an automated system for restoring blurred images by the Lucy - Richardson method is directly connected with determining the optimal number of iterations of this method to obtain the best quality image. It is preferable to use reference quality measures as a criterion for stopping the iterative process than reference-free measures because they are more strongly correlated with the image quality perceived by a human. However, in practice, only distorted images are available to automated restoration systems. In this work, an approach for determining the number of iterations for the Lucy - Richardson method is proposed based on predicting the optimal number of iterations of the PieAPP reference measure using the CS reference-free measure. The key recovery problem, the estimation of the distorting operator, has been solved using a neural network algorithm based on the ideas of autoencoders and the Xception neural network architecture. It was demonstrated that the use of the proposed approach allows the improvement of the reconstructed image quality for the PieAPP reference measure compared with a reconstruction scenario with a fixed number of iterations. Thus, the quality of automated (with no operator participation) reconstruction of distorted images by the Lucy - Richardson method can be improved.

Key words: image restoration, blur compensation, Lucy – Richardson method, automated systems, neural networks
Published in: INFORMATION-COMMUNICATION TECHNOLOGIES
Bibliography link: Bordiuzha V., Breykina K. V., Umnyashkin S. V. Automation of determining the number of iterations in the problem of blurred images restoration by the Lucy – Richardson method. Proc. Univ. Electronics, 2024, vol. 29, no. 4, pp. 539–550. https://doi.org/10.24151/1561-5405-2024-29-4-539-550

Viktor Bordiuzha
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)

Kristina V. Breykina
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)

Sergey V. Umnyashkin
National Research University of Electronic Technology (Russia, 124498, Moscow, Zelenograd, Shokin sq., 1)

1. Гонсалес Р. С., Вудс Р. E. Цифровая обработка изображений. 3-е изд., испр. и доп. М.: Техносфера, 2012. 1104 с. EDN: TIKLUW
Gonzalez R. C., Woods R. E. Digital Image Processing. 3rd ed. Upper Saddle River, NJ, Pearson, 2007. 976 p.

2. Умняшкин С. В. Основы теории цифровой обработки сигналов: учеб. пособие. 6-е изд., испр. М.: Техносфера, 2021. 550 с. EDN: XJXXWU
Umnyashkin S. V. Fundamentals of the theory of digital signal processing, study guide. 6th ed., rev. Moscow, Tekhnosfera Publ., 2021. 550 p. (In Russian).

3. Biemond J., Lagendijk R. L., Mersereau R. M. Iterative methods for image deblurring // Proceedings of the IEEE. 1990. Vol. 78. No. 5. P. 856-883. DOI: 10.1109/5.53403

4. Бордюжа В., Умняшкин С. В. Предобработка изображений для повышения качества распознавания лиц на цифровых фото // Цифровая обработка сигналов и ее применение (DSPA-2023): доклады XXV Междунар. конф. (Москва, 29-31 мар. 2023). М.: Рос. НТО радиотехники, электрон. и связи им. А. С. Попова, 2023. С. 299-304. EDN: BSVLHJ
Bordiuzha V., Umnyashkin S. V. Image preprocessing to improve facial recognition in digital photos. Tsifrovaya obrabotka signalov i eye primeneniye (DSPA-2023), proceedings of 25th International conf. (Moscow, March 29–31, 2023). Moscow, Ros. NTO radiotekhniki, elektron. i svyazi im. A. S. Popova Publ., 2023, pp. 299–304. (In Russian).

5. Lucy L. B. An iterative technique for the rectification of observed distributions // Astron. J. 1974. Vol. 79. No. 6. P. 745-754. DOI: 10.1086/111605

6. Bank D., Koenigstein N., Giryes R. Autoencoders // Machine Learning for Data Science Handbook / eds L. Rokach et al. Cham: Springer International, 2023. P. 353-374. DOI: 10.1007/978-3-031-24628-9_16

7. Chollet F. Xception: Deep learning with depthwise separable convolutions // ArXiv [Электронный ресурс]. 07.10.2016. 10.48550/arXiv.1610.02357 (дата обращения: 20.05.2024). DOI: 10.48550/arXiv.1610.02357

8. Statistics for the evaluation and comparison of models / C. J. Willmott, S. G. Ackleson, R. E. Davis et al. //j. Geophys. Res. 1985. Vol. 90. Iss. C5. P. 8995-9005. DOI: 10.1029/JC090iC05p08995

9. Eskicioglu A. M., Fisher P. S. Image quality measures and their performance // IEEE Transactions on Communications. 1995. Vol. 43. No. 12. P. 2959-2965. DOI: 10.1109/26.477498

10. Wang Z., Bovik A. C., Sheikh H. R., Simoncelli E. P. Image quality assessment: From error visibility to structural similarity // IEEE Transactions on Image Processing. 2004. Vol. 13. No. 4. P. 600-612. DOI: 10.1109/TIP.2003.819861

11. Prashnani E., Cai H., Mostofi Y., Sen P. PieAPP: Perceptual image-error assessment through pairwise preference // 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Salt Lake City, UT: IEEE, 2018. P. 1808-1817. DOI: 10.1109/CVPR.2018.00194 EDN: XANJBV

12. Mittal A., Moorthy A. K., Bovik A. C. No-reference image quality assessment in the spatial domain // IEEE Transactions on Image Processing. 2012. Vol. 21. No. 12. P. 4695-4708. DOI: 10.1109/TIP.2012.2214050

13. Mittal A., Soundararajan R., Bovik A. C. Making a "completely blind" image quality analyzer // IEEE Signal Processing Letters. 2013. Vol. 20. No. 3. P. 209-212. DOI: 10.1109/LSP.2012.2227726

14. Smola A. J., Schölkopf B. A tutorial on support vector regression // Statistics and Computing. 2004. Vol. 14. Iss. 3. P. 199-222. https://doi.org/10.1023/B:STCO.0000035301.49549.88. EDN: DCRZKF

15. Брейкина К. В., Умняшкин С. В. Оценка качества изображения при компенсации смаза по методу Люси - Ричардсона // Изв. вузов. Электроника. 2020. Т. 25. № 2. С. 167-174. DOI: 10.24151/1561-5405-2020-25-2-167-174 EDN: GDDQFI
Breykina K. V., Umnyashkin S. V. Image quality estimation for blur compensation using Lucy – Richardson method. Izv. vuzov. Elektronika = Proc. Univ. Electronics, 2020, vol. 25, no. 2, pp. 167–174. (In Russian). https://doi.org/10.24151/1561-5405-2020-25-2-167-174

16. Mei C., Reid I. Modeling and generating complex motion blur for real-time tracking // 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK: IEEE, 2008. P. 1-8. DOI: 10.1109/CVPR.2008.4587535

17. Scikit-learn: Machine learning in Python // Scikit-Learn [Электронный ресурс]. URL: https://scikit-learn.org/stable (дата обращения: 20.05.2024).

Information about the publication

Downloads: 0

UDK: 004.932.4
Publication type: Scientific article
DOI: 10.24151/1561-5405-2024-29-4-539-550
RISC id: GUFPVG

Download: File not found (680.58 Kb) JATS XML