RESEARCH OF THE TEXTURES SYNTHESIS METHOD OT THE SURFACE IMAGE OF THE EARTH BASED ON A NEURAL NETWORK
Abstract
The removal of cloud images from photographs of the underlying surface is a prerequisite for the use of datasets from Earth satellites, since such satellite images are used to analyze chang-es in soil cover associated with urban expansion, restoration of forests on abandoned agricultural lands, estimation of forest area, identification of forest types and classification land cover, and this data is most often polluted by clouds. The article explores the algorithm for extracting and remov-ing cloud images and develops its software implementation. The effectiveness of the new approach is shown using several examples for various areas of the earth's surface with clouds. The subject of research is the methods and algorithms for detecting and reconstructing objects that hide de-tails on images, in particular cloud images. The object of study is a set of test images. The result of the study is the development of a method for removing cloud images in order to restore the area covered by clouds. The novelty of the work is an algorithm that improves the quality of image re-covery based on a neural network. The results obtained make it possible to reconstruct areas cov-ered by clouds. The effectiveness of the image restoration method was evaluated using a statistical criterion – the random mean square error of the processing result from the true image. As a result of solving the tasks, it can be concluded: – a method has been developed for removing cloud imag-es and restoring images based on the search for similar blocks with their subsequent integration by a neural network; – an analysis of the results of the study showed that the proposed method can improve the quality of image reconstruction.
References
7. Lysenko T.M. An Adaptive Algorithm of Edge Detection in Radar Images, 2015 25rd Interna-tional Crimean Conference «Microwave & Telecommunication Technology» (CriMiCo, 2015). September 6-12, 2015. Sevastopol, Conference Proceedings. In Two Volumes – Sevastopol, 2015, Vol. 2, pp. 1234-1235. 8. De I., Chanda B. A simple and efficient algorithm for multifocus image fusion using morpho-logical wavelets, Signal Processing, 2006, Vol. 86, pp. 924-936. 9. Gomathi R., Kumar A. Inpainting for satellite imagery using thin plate spline radial basis func-tion neural networks in shearlet domain, Journal of Intelligent & Fuzzy Systems, 2014, Vol. 27, No. 5, pp. 2391-2398. 10. Shovengerdt R.A. Distantsionnoe zondirovanie. Modeli i metody obrabotki izobrazheniy [Re-mote sensing. Models and methods of image processing]. Moscow: Tekhnosfera, 2010, 560 p. 11. Kobernichenko V.G. Radioelektronnye sistemy distantsionnogo zondirovaniya Zemli [Radio-electronic systems of remote sensing of the Earth]. Ekaterinburg: Izd-vo Ural'skogo universiteta, 2016, 220 p. 12. Shen H., Zhang L. A MAP-based algorithm for destriping and inpainting of remotely sensed images, IEEE Transactions on Geoscience and Remote Sensing, 2008, Vol. 47, No. 5, pp. 1492-1502. 13. Kobernichenko V.G., Trenikhin V.A. Analiz i razrabotka metodov povysheniya informativnosti kosmicheskikh radiolokatsionnykh i mul'tispektral'nykh izobrazheniy [Analysis and develop-ment of methods for increasing the information content of space radar and multispectral imag-es], Nauchnye trudy XII otchetnoy konferentsii molodykh uchenykh UGTU-UPI [Scientific works of the XII reporting conference of young scientists of USTU-UPI], 2007, pp. 227-230. 14. Semenishchev E.A., Voronin V.V., Marchuk V.I. Image denoising and deblurring using multi-spectral data, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXIII. International Society for Optics and Photonics, 2017, Vol. 10198, pp. 101981G. 15. Bertalmio M., Bertozzi A.L., Sapiro G. Navier-stokes, fluid dynamics, and image and video inpainting, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001. IEEE, 2001, Vol. 1, pp. I-I. 16. Guleryuz O.G. Nonlinear approximation based image recovery using adaptive sparse recon-structions and iterated denoising-part I: theory, IEEE Transactions on image processing, 2006, Vol. 15, No. 3, pp. 539-554.
17. Ibadov R.R., Ibadov S.R., Voronin V.V., Fedosov V.P. Modifitsirovannyy metod rekonstruktsii izobrazheniy na osnove poiska podobnykh oblastey [A modified method of image reconstruc-tion based on the search for such areas], Izvestiya YuFU. Tekhnicheskie nauki [Izvestiya SFedU. Engineering Sciences], 2017, No. 6 (191), pp. 179-189.
18. Voronin V.V., Marchuk V.I., Sherstobitov A.I., Semenishchev E.A., Frantc V.A. Image recon-struction on the basis of a textural geometrical model, Pattern Recognition and Image Analy-sis, 2015, Vol. 25, No. 3, pp. 553-562.
19. Fedosov V.P., Ibadov R.R., Ibadov S.R., Voronin V.V. Restoration of the Blind Zone of the Image of the Underlying Surface for Radar Systems with Doppler Beam Sharpening, 2019 Radiation and Scattering of Electromagnetic Waves (RSEMW). IEEE, 2019, pp. 424-427. 20. Gapon N., Ponamorenko M., Pismenskova M., Tokareva O. Image inpainting using a neural network, MATEC Web of Conferences. EDP Sciences, 2017, Vol. 132, pp. 05015.
