Журнал

About Editorial Board Information for authors and reviewers Publication ethics Contacts User Account: send / review a manuscript
Archive
Volume 21, 2024
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 20, 2023
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 19, 2022
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 18, 2021
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 17, 2020
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 16, 2019
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 15, 2018
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 14, 2017
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6 Number 7
Volume 13, 2016
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 12, 2015
Number 1 Number 2 Number 3 Number 4 Number 5 Number 6
Volume 11, 2014
Number 1 Number 2 Number 3 Number 4
Volume 10, 2013
Number 1 Number 2 Number 3 Number 4
Volume 9, 2012
Number 1 Number 2 Number 3 Number 4 Number 5
Volume 8, 2011
Number 1 Number 2 Number 3 Number 4
Volume 7, 2010
Number 1 Number 2 Number 3 Number 4
Issue 6, 2009
Volume 1 Volume 2
Issue 5, 2008
Volume 1 Volume 2
Issue 4, 2007
Volume 1 Volume 2
Issue 3, 2006
Volume 1 Volume 2
Issue 2, 2005
Volume 1 Volume 2
Issue 1, 2004
Volume 1
Search
Find:
русский
ISSN 2070-7401 (Print), ISSN 2411-0280 (Online)
Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa
CURRENT PROBLEMS IN REMOTE SENSING OF THE EARTH FROM SPACE

  

Sovremennye problemy distantsionnogo zondirovaniya Zemli iz kosmosa, 2024, Vol. 21, No. 2, pp. 36-50

Cloud and snow neural network segmentation using Electro-L No. 2 satellite multispectral data

N.V. Belyakov 1 , A.V. Vasiliev 2 , S.V. Kolpinskiy 3 
1 Lomonosov Moscow State University, Moscow, Russia
2 Research Computing Center of Lomonosov Moscow State University, Moscow, Russia
3 National Research University "Moscow Power Engineering Institute", Moscow, Russia
Accepted: 29.03.2024
DOI: 10.21046/2070-7401-2024-21-2-36-50
This paper is devoted to cloud and snow semantic segmentation of multispectral satellite imagery of a multizone scanning geostationary instrument (MSU-GS) installed on the Russian satellite Electro-L No. 2 using a convolutional neural network. To develop the neural network algorithm, a self-collected dataset containing multispectral images from the MSU-GS instrument installed on the Electro-L No. 2 satellite with snow and cloud masks was created. The images from GOES-16 (Geostationary Operational Environmental Satellite) and Meteosat-10 satellites reprojected to the Electro-L No. 2 position were used to create cloud and snow cover masks. Such geographical information as latitude, longitude and altitude for image pixels is used as additional information. Multi-Scale Attention Network (MANet) is used as a neural network model. One of the significant problems in the development of the snow and cloud segmentation algorithm is the absence of short-wave infrared channels (1300–1600 nm), which are necessary for the operation of classical segmentation algorithms based on normalized snow index tests. Given the limitations on the spectral characteristics of the MSU-GS equipment and low resolution of images, a neural network algorithm capable of differentiating snow and clouds by implicit features and patterns is proposed as a solution to the problem of snow and cloud separation on multispectral data. For representativeness, the images in the dataset include all seasons and different light levels (12:00–17:00 UTC). The trained neural network for segmentation was tested in different scenarios, including winter and summer periods of the year in daytime at different illumination levels of the images on the data from the MSU-GS instrument of the Electro-L No. 2 satellite. According to the test results the following segmentation quality metrics were obtained: F1s = 0,7454, F1c = 0,8773 and IoUs = 0,7398, IoUc = 0,7976 (Intersection over Union index) for snow (subscript s) and cloud (subscript c) classes respectively.
Keywords: Electro-L No. 2, remote sensing, convolutional neural network, cloud and snow segmentation, geographical information, altitude map
Full text

References:

  1. Astafurov V. G., Evsyutkin T. V., Kuriyanovich K. V., Skorokhodov A. V., Statistical model of cirrus cloud textural features based on MODIS satellite images, Atmospheric and Oceanic Optics, 2014, Vol. 27, No. 7, pp. 640–646 (in Russian).
  2. Ackerman S. A., Frey R., MODIS Atmosphere L2 Cloud Mask Product, NASA MODIS Adaptive Processing System, Goddard Space Flight Center, 2015, DOI: 10.5067/MODIS/MOD35_L2.061.
  3. Bloshchinskiy V. D., Kuchma M., Andreev A., Sorokin A., Snow and cloud detection using a convolutional neural network and low-resolution data from the Electro-L No. 2 Satellite, J. Applied Remote Sensing, 2020, Vol. 14, No. 3, Article 034506, DOI: 10.1117/1.JRS.14.034506.
  4. Cao H., Luo X., Peng Y. et al., MANet: A Network Architecture for Remote Sensing Spatiotemporal Fusion Based on Multiscale and Attention Mechanisms, Remote Sensing, 2022, Vol. 14, No. 18, pp. 4600–4600, https://doi.org/10.3390/rs14184600.
  5. Fan T., Wang G., Li Y. et al., MA-Net: A Multi-Scale Attention Network for Liver and Tumor Segmentation, IEEE Access, 2020, Vol. 8, pp. 179656–179665, DOI: 10.1109/ACCESS.2020.3025372.
  6. Heidinger A., Botambekov D., Walthe A., Algorithm theoretical basis document, A Naïve Bayesian Cloud Mask Delivered to NOAA Enterprise, 10 July, 2017, 50 p., https://www.ospo.noaa.gov/Products/Suites/files/atbd/Cloud_Mask_Enterprise_ATBD_v1.1_2016.pdf.
  7. Khlopenkov K., Trishchenko A., SPARC: New cloud, snow, and cloud shadow detection scheme for historical 1-km AVHHR data over Canada, J. Atmospheric and Oceanic Technology, 2007, Vol. 24, No. 3, pp. 322–343, DOI: 10.1175/JTECH1987.1.
  8. Krizhevsky A., Sutskever I., Hinton G. E., ImageNet Classification with Deep Convolutional Neural Networks, arXiv preprint, arXiv:1404.5997, 2012, http://arxiv.org/abs/1404.5997.
  9. Li R., Zheng S., Duan C. et al., Multiattention Network for Semantic Segmentation of Fine Resolution Remote Sensing Images, IEEE Trans. Geoscience and Remote Sensing, 2022, Vol. 60, Article 5607713, pp. 1–13, DOI: 10.1109/TGRS.2021.3093977.
  10. Lin T., Goyal P., Girshick R., He K., Dollar P., Focal Loss for Dense Object Detection, arXiv preprint, arXiv:1708.02002, 2018, https://arxiv.org/pdf/arXiv.1708.02002.
  11. Loshchilov I., Hutter F., SGDR: Stochastic Gradient Descent with Warm Restarts, arXiv preprint, arXiv:1608.03983v5, 2017, https://doi.org/10.48550/arXiv.1608.03983.
  12. Loshchilov I., Hutter F., Decoupled weight decay regularization, arXiv preprint, arXiv:1711.05101, 2019, https://arxiv.org/pdf/1711.05101.
  13. Miller S., Lee T., Fennimore R. L., Satellite-based imagery techniques for daytime cloud/snow delineation from MODIS, J. Applied Meteorology, 2005, Vol. 44, No. 7, pp. 987–997, DOI: 10.1175/JAM2252.1.
  14. Mohajerani S., Saeedi P., Cloud-Net: An End-To-End Cloud Detection Algorithm for Landsat 8 Imagery, 2019 IEEE Intern. Geoscience and Remote Sensing Symp. (IGARSS 2019), 2019, pp. 1029–1032.
  15. Sun S., Cao Z., Zhu H., Zhao J., A Survey of Optimization Methods From a Machine Learning Perspective, arXiv preprint, arXiv:1906.06821, 2019, https://arxiv.org/pdf/arXiv.1906.06821.
  16. Tan M., Le Q. V., EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (Image Classification), arXiv preprint, arXiv:1905.11946, 2019, https://doi.org/10.48550/arXiv. 1905.11946.
  17. Wildt M. D. R. D., Seiz G., Gruen A., Operational snow mapping using multitemporal Meteosat SEVIRI imagery, Remote Sensing of Environment, Elsevier Inc., 2007, Vol. 109, pp. 29–41, DOI: 10.1016/j.rse.2006.12.008.
  18. Wortsman M., Ilharco G., Gadre S. Y. et al., Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time, Proc. 39 th Intern. Conf. Machine Learning. PMLR, 2022, Vol. 162, pp. 23965–23998, https://doi.org/10.48550/arXiv.2203.05482.
  19. Wright L., Demeure N., Ranger21: a synergistic deep learning optimizer, arXiv preprint, arXiv:2106.13731, 2021, https://doi.org/10.48550/arXiv.2106.13731.
  20. Wu X., Shi Z., Zou Z., A geographic information-driven method and a new large scale dataset for remote sensing cloud/snow detection, ISPRS J. Photogrammetry and Remote Sensing, 2021, Vol. 174, No. 11, pp. 87–104, DOI: 10.1016/j.isprsjprs.2021.01.023.
  21. Wu Y., Liu L., Bae J. et al., Demystifying Learning Rate Policies for High Accuracy Training of Deep Neural Networks, IEEE Intern. Conf. Big Data, 2019, pp. 1971–1980, DOI: 10.1109/BigData47090.2019.9006104.
  22. Yin M., Wang P., Ni C. et al., Cloud and snow detection of remote sensing images based on improved Unet3+, Scientific Reports, 2022, Vol. 12. No. 1, DOI: 10.1038/s41598-022-18812-6.
  23. Zhang J., Li X., Li L. et al., Lightweight U-Net for cloud detection of visible and thermal infrared remote sensing images, Optical and Quantum Electronics, 2020, Vol. 52, No. 9, pp. 397–397, DOI: 10.1007/s11082-020-02500-8.