Elimination of cloud shadows on materials of aviation shooting in the visible range


View or download the article (Rus)

About the Authors

Ilya A. Rylskiу

Moscow State University named after M.V. Lomonosov, Faculty of Geography,
Moscow, 119991, Russia,
E-mail: rilskiy@mail.ru

Evgeniy N. Eremchenko

Moscow State University named after M.V. Lomonosov, Faculty of Geography,
Moscow, 119991, Russia,
E-mail: eugene.eremchenko@gmail.com

Tatiana V. Kotova

Moscow State University named after M.V. Lomonosov, Faculty of Geography,
Moscow, 119991, Russia,
E-mail: tatianav.kotova@yandex.ru


Aerial photography is often impossible due to the presence of high clouds with contrasting shadows that do not allow to obtain materials suitable for decryption. At the same time, in a significant proportion of projects in Russia, the snowless season suitable for surveying is very short. The inability to perform aerial photography while flying below the clouds leads to cost increasing. In some cases, projects cannot be completed.

Existing software does not allow to solve the problem of equalizing the brightness in the shadows for several reasons. The main reason is the inability to identify the boundaries of the shadows using only the spectral characteristics of the images, the inability to determine the amount of correction for shaded areas.

To solve this problem, it is proposed to use reference images of the worse resolution obtained from the satellites. Reference images are used to localize and determine the magnitude of the spectral correction of aerial photographs. The work is performed with single orthophotographs or orthophotomosaics in the same coordinate system. To determine the boundaries of the shaded zones and the values of the corrections in brightness, methods of cartographic algebra on regular data arrays are used. Further, the obtained correction matrices are subject to filtering and are used to correct high-resolution aerial photographs.

The paper gives an example of the use of free (or cheap) satellite images to eliminate or reduce the contrast of shadows on aerial photographs with a detail of 20 cm. The created prototype software allows to perform additive or multiplicative correction of an array of individual aerial photographs.

The proposed approach requires more time for data processing, but gives much more acceptable results for visual (manual) decryption. The method is not recommended for use when working with images in more than 10 cm, when solving monitoring tasks with frequent repeated surveys, and also, if necessary, to carry out automated decoding using spectral standards.


airborne imagery, satellite images, remote sensing, GIS, geoinformation data


  1. Dare M. Shadow analysis in high-resolution satellite imagery of urban areas. Photogrammetric Engineering Remote Sensing, 2005. V. 71. Р. 169–177. DOI: 10.14358/PERS.71.2.169.
  2. Finlayson G., Drew M., Lu C. Entropy minimization for shadow removal. International Journal of Computer Vision, 2009. No 85. P. 35–57.
  3. Finlayson G.D., Hordley S.D., Drew M.S. Removing shadows from images. 7th European Conference on Computer Vision. Berlin, Heidelberg: Springer, 2002. P. 823–836.
  4. Guidance in aerial photography. Moscow: Ministry of Civil Aviation, 1986. 176 p. (in Russian).
  5. Guidance in aerial photography for cartographic purposes. Military Topographical Directorate of the General Staff. Moscow: Printing and publication department, 1989. 105 p. (in Russian).
  6. Guo R., Dai Q., Hoiem D. Single-image shadow detection and removal using paired regions. CVPR (The Conference on Computer Vision and Pattern Recognition), Colorado Springs. IEEE, 2011. Р. 2033–2040.
  7. Kapralov E.G., Koshkarev A.V., Tikunov V.S. Fundamentals of geoinformatics. Мoscow: Academy, 2004. 480 p.
  8. Levine M.D., Bhattacharyya J. Removing shadows. Pattern Recognition Letters, 2005. V. 26. Iss. 3. Р. 251–265. DOI: 10.1016/j.patrec.2004.10.021.
  9. Liu F., Gleicher M. Texture-consistent shadow removal. ECCV (10th European Conference on Computer Vision), Marseille, France, 2008. Lectures Notes in Computer Science. Berlin, Heidelberg: Springer-Verlag, 2008. V. 5305. P. 437–450. DOI: https://doi.org/10.1007/978-3-540-88693-8_32.
  10. Shor Y., Lischinski D. The shadow meets the mask: Pyramid-based shadow removal. Computer Graphics Forum, 2008. V. 27. Iss. 2. P. 577–586. DOI: https://doi.org/10.1111/j.1467-8659.2008.01155.x.
  11. Zhu J., Samuel K., Masood S.Z., Tappen M. Learning to recognize shadows in monochromatic natural images. CVPR (IEEE Computer Society Conference on Computer Vision and Pattern Recognition), San-Francisco, CA, USA, 2010. Proceedings. P. 223–230. DOI: 10.1109/CVPR.2010.5540209.

For citation: Rylskiу I.A., Eremchenko E.N., Kotova T.V. Elimination of cloud shadows on materials of aviation shooting in the visible range InterCarto. InterGIS. GI support of sustainable development of territories: Proceedings of the International conference. Moscow: Moscow University Press, 2020. V. 26. Part 2. P. 286–297. DOI: 10.35595/2414-9179-2020-2-26-286-297 (In Russian)