Skip to main content

Advertisement

SpringerLink
Log in

Classification and Understanding of Cloud Structures via Satellite Images with EfficientUNet

  • Original Research
  • Published:
SN Computer Science Aims and scope Submit manuscript

Abstract

Climate change has been a common interest and the forefront of crucial political discussion and decision-making for many years. Shallow/Low-altitude clouds play a significant role in understanding the Earth’s climate, but they are challenging to interpret and represent in a climate model. By classifying these cloud structures, there is a better possibility of understanding the physical structures of the clouds, which would improve the climate model generation, resulting in a better prediction of climate change or forecasting weather update. Clouds organise in many forms, which makes it challenging to build traditional rule-based algorithms to separate cloud features. In this paper, classification of cloud organization patterns was performed using a new scaled-up version of Convolutional Neural Network (CNN) named as EfficientNet as the encoder and UNet as decoder where they worked as feature extractor and reconstructor of fine grained feature map and was used as a classifier, which will help experts to understand how clouds will shape the future climate. By using a segmentation model in a classification task, it was shown that with a good encoder alongside UNet, it is possible to obtain good performance from this dataset. Dice coefficient has been used for the final evaluation metric, which gave the score of 66.26% and 66.02% for public and private (test set) leaderboard on Kaggle competition respectively.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9

Similar content being viewed by others

Notes

  1. LB: Leaderboard.

References

  1. Agarap AF. Deep learning using rectified linear units (relu). 2018; arXiv preprint.

  2. B. Baheti, S. Innani, S. Gajre and S. Talbar, "Eff-UNet: A Novel Architecture for Semantic Segmentation in Unstructured Environment," in 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Seattle, WA, USA, 2020;1:1473–1481.

  3. Buslaev A, Iglovikov VI, Khvedchenya E, Parinov A, Druzhinin M, Kalinin AA. Albumentations: fast and flexible image augmentations. Information. 2020;11(2):125.

    Article  Google Scholar 

  4. de Lima GRT, Stephany S. A new classification approach for detecting severe weather patterns. Comput Geosci. 2013;57:158–65.

    Article  Google Scholar 

  5. Dice LR. Measures of the amount of ecologic association between species. Ecology. 1945;26(3):297–302.

    Article  Google Scholar 

  6. Dowling DR, Radke LF. A summary of the physical properties of cirrus clouds. J Appl Meteorol. 1990;29(9):970–8.

    Article  Google Scholar 

  7. Esch T, Thiel M, Bock M, Roth A, Dech S. Improvement of image segmentation accuracy based on multiscale optimization procedure. IEEE Geosci Remote Sens Lett. 2008;5(3):463–7.

    Article  Google Scholar 

  8. Gadsden M, Schröder W. Noctilucent clouds. In: Noctilucent Clouds. Springer; 1989, p. 1–12.

  9. He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on computer vision and pattern recognition, 2016;1:770–78.

  10. Laban N, Abdellatif B, Ebied HM, Shedeed HA, Tolba MF. Multiscale satellite image classification using deep learning approach. In: Machine learning and data mining in aerospace technology. Springer; 2020, p. 165–186.

  11. Leconte J, Forget F, Charnay B, Wordsworth R, Pottier A. Increased insolation threshold for runaway greenhouse processes on earth-like planets. Nature. 2013;504(7479):268–71.

    Article  Google Scholar 

  12. Lee J, Won T, Hong K. Compounding the performance improvements of assembled techniques in a convolutional neural network. 2020; arXiv preprint arXiv:2001.06268.

  13. Liu L, Jiang H, He P, Chen W, Liu X, Gao J, Han J. On the variance of the adaptive learning rate and beyond. 2019; arXiv preprint arXiv:1908.03265.

  14. Long J, Shelhamer E, Darrell T. Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Conference on computer vision and pattern recognition, 2015;1:3431–440.

  15. Mason J, Strapp W, Chow P. The ice particle threat to engines in flight. In: 44th AIAA Aerospace Sciences Meeting and Exhibit, 2006; p. 206.

  16. McLean GS. Cloud distributions in the vicinity of jet streams. Bull Am Meteorol Soc. 1957;38(10):579–83.

    Article  Google Scholar 

  17. Miller J, Nair U, Ramachandran R, Maskey M. Detection of transverse cirrus bands in satellite imagery using deep learning. Comput Geosci. 2018;118:79–85.

    Article  Google Scholar 

  18. Niwattanakul S, Singthongchai J, Naenudorn E, Wanapu S. Using of jaccard coefficient for keywords similarity. In: Proceedings of the International Multiconference of engineers and computer scientists. 2013;1:380–4.

  19. Pallé E, Goode PR, Montanes-Rodriguez P, Koonin SE. Changes in earth’s reflectance over the past two decades. Science. 2004;304(5675):1299–301.

    Article  Google Scholar 

  20. Pritt M, Chern G. Satellite image classification with deep learning. In: 2017 IEEE Applied Imagery Pattern Recognition Workshop (AIPR), 2017;1:1–7. IEEE.

  21. Richards ND. Method of encoding image pixel values for storage as compressed digital data and method of decoding the compressed digital data, December 14 1993. US Patent 5270812.

  22. Ronneberger O, Fischer P, Brox T. U-net: Convolutional networks for biomedical image segmentation. In: International Conference on medical image computing and computer-assisted intervention, 2015;9351:234–241. Springer.

  23. Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L-C. Mobilenetv2: Inverted residuals and linear bottlenecks. In: Proceedings of the IEEE Conference on computer vision and pattern recognition, 2018;1:4510–520.

  24. Sudre CH, Li W, Vercauteren T, Ourselin S, Jorge Cardoso M. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In: Deep learning in medical image analysis and multimodal learning for clinical decision support. Springer; 2017, p. 240–248.

  25. Tan M, Le QV. Efficientnet: rethinking model scaling for convolutional neural networks. 2019; arXiv preprintarXiv:1905.11946..

  26. Turner DD, Vogelmann AM, Austin RT, Barnard JC, Cady-Pereira K, Christine Chiu J, Clough SA, Flynn C, Khaiyer MM, Liljegren J, et al. Thin liquid water clouds: Their importance and our challenge. Bull Am Meteorol Soc. 2007;88(2):177–90.

    Article  Google Scholar 

  27. Twomey S, et al. Pollution and the planetary albedo. Atmos Environ. 1974;8(12):1251–6.

    Article  Google Scholar 

  28. Wang J, Cohen MF. An iterative optimization approach for unified image segmentation and matting. In: Tenth IEEE International Conference on computer vision (ICCV’05) Volume 1, volume 2, pages 936–943. IEEE, 2005.

  29. Xia M, Tian N, Zhang Y, Yiqing X, Zhang X. Dilated multi-scale cascade forest for satellite image classification. Int J Remote Sens. 2020;41(20):7779–800.

    Article  Google Scholar 

  30. Yakubovskiy P. Segmentation models. 2019; https://github.com/qubvel/segmentation_models .

  31. Zhang Q, Quan J, Tie X, Huang M, Ma X. Impact of aerosol particles on cloud formation: aircraft measurements in china. Atmos Environ. 2011;45(3):665–72.

    Article  Google Scholar 

Download references

Acknowledgements

We thank Kaggle for hosting such a great contest and for their kernel (notebook). We also thank Max Planck Institute for Meteorology for providing the dataset. We also admire Pavel Yakubovskiy for sharing his contribution in GitHub repository from where we managed to access the pretrained imagenet weights. Authors would like to thank the anonymous reviewers for providing suggestions which improved the quality of this article. Source code of this project is available at .

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Author information

Authors and Affiliations

  1. United International University, Dhaka, Bangladesh

    Tashin Ahmed & Noor Hossain Nuri Sabab

Authors
  1. Tashin Ahmed
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Noor Hossain Nuri Sabab
    View author publications

    You can also search for this author in PubMed Google Scholar

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ahmed, T., Sabab, N.H.N. Classification and Understanding of Cloud Structures via Satellite Images with EfficientUNet. SN COMPUT. SCI. 3, 99 (2022). https://doi.org/10.1007/s42979-021-00981-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s42979-021-00981-2

Keywords

Search

Navigation