VKA named after A. F. Mozhaisky (Department of Mathematics and Software, Professor)
Russian Federation
Russian Federation
Russian Federation
Abstract. The problem of processing semi-structured and unstructured Earth remote sensing (ERS) data obtained by various methods, including satellites and unmanned aerial vehicles, is touched upon. The purpose of the article is to study and implement algorithms for effective purification and preprocessing of semi-structured and unstructured Earth remote sensing data. The importance of cleaning these data from noise, artifacts and errors is emphasized in order to increase their accuracy and significance in applied scientific research and practical application. Key data preprocessing techniques are considered, including noise removal, distortion correction, classification, segmentation and standardization of data, reinforcing theoretical positions with practical examples in Python using libraries such as GDAL, OpenCV and scikit-image. Examples of detection of aerospace and transport objects using machine learning and deep learning are given, emphasizing the importance of accuracy, completeness and F1-Score metrics in assessing the quality of data purification. The practical significance of the study lies in evaluating the effectiveness of data purification methods used to restore images during remote sensing of the Earth.
remote sensing of the Earth, data preprocessing, data purification, machine learning, deep learning, noise filtering, distortion correction, data classification, image segmentation, aerospace objects, GDAL, OpenCV, scikit-image
1. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising / K. Zhang [et al.] // IEEE Transactions on Image Processing. 2017. P. 3142–3155.
2. Komp'yuternoe zrenie. Sovremennye metody i perspektivy razvitiya / red. R. Devis, M. Terk; per. s angl. V. S. Yacenkova. M.: DMK Press, 2022. 690 s.
3. Nikitin G. Yu. Povyshenie kachestva izobrazheniya na baze algoritmov neyronnyh setey // Komp'yuternye sistemy i seti: materialy 54-y nauchnoy konferencii aspirantov, magistrantov i studentov (Minsk, Belarus', 23–27 aprelya 2018 g.). Minsk: Belorusskiy gos. un-t informatiki i radioelektroniki, 2018. S. 242–244.
4. Bulyga F. S., Kureychik V. M. Metod ponizheniya shuma na cifrovyh izobrazheniyah // Mirovye nauchnye issledovaniya i razrabotki v epohu cifrovizacii: materialy XV Mezhdunarodnoy nauchno-prakticheskoy konferencii (Rostov-na-Donu, Rossiya, 25 noyabrya 2021 g.): v 2 ch. Ch. 1. Rostov n/D.: Izd-vo Yuzhnogo universiteta (IUBiP), 2021. S. 143–147.
5. Kislyanskiy G. N., Nestrugina E. S. Vosstanovlenie rasfokusirovannyh i smazannyh izobrazheniy // Vestnik Doneckogo nacional'nogo universiteta. Seriya G: Tehnicheskie nauki. 2020. № 4. S. 41–53.
6. Learning to Estimate and Remove Non-uniform Image Blur / F. Couzinié-Devy [et al.] // Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR 2013) (Portland, OR, USA, 23–28 June 2013). Institute of Electrical and Electronics Engineers, 2013. P. 1075–1082. DOI:https://doi.org/10.1109/CVPR.2013.143
7. Vershovskiy E. A. Roevoy algoritm optimizacii v zadache klasterizacii mul'tispektral'nogo snimka // Izvestiya Yuzhnogo federal'nogo universiteta. Tehnicheskie nauki. 2010. № 5(106). S. 102–107.
8. Vershovskiy E. A. Metod kontrolya kachestva klasterizacii mul'tispektral'nogo izobrazheniya // Izvestiya Yuzhnogo federal'nogo universiteta. Seriya «Tehnicheskie nauki». 2010. № 7(108). S. 191–198.
9. Kuznecov A. A., Oparin A. N., Shishkin V. A. Poisk i raspoznavanie ob'ektov na bazise neyrosetevyh algoritmov i neyroprocessornyh tehnologiy // Prikladnaya fizika. 2006. № 5. S. 97–100.
10. Sorokin S. V. Vozmozhnosti i preimuschestva vzveshennyh mediannyh fil'trov dlya udaleniya impul'snogo shuma na izobrazhenii // Trudy XIX Mezhdunarodnogo simpoziuma «Nadezhnost' i kachestvo» (Penza, Rossiya, 26 maya — 01 iyunya 2014 g.). T. 2. Penza: Penzenskiy gos. un-t, 2014. S. 203–204.
