Evaluation of Cloud Mask Performance of KOMPSAT-3 Top-of-Atmosphere Reflectance Incorporating Deeplabv3+ with Resnet 101 Model
Source abstract
Abstract
Cloud detection is a crucial task in satellite remote sensing, influencing applications such as vegetation indices, land use analysis, and renewable energy estimation. This study evaluates the performance of cloud masks generated for KOMPSAT-3 and KOMPSAT-3A imagery using the DeepLabV3+ deep learning model with a ResNet-101 backbone. To overcome the limitations of digital number (DN) data, Top-of-Atmosphere (TOA) reflectance was computed and used for model training. Comparative analysis between the DN and TOA reflectance demonstrated significant improvements with the TOA correction applied. The TOA reflectance combined with the NDVI channel achieved the highest precision (69.33%) and F1-score (59.27%), along with a mean Intersection over Union (mIoU) of 46.5%, outperforming all the other configurations. In particular, this combination was highly effective in detecting dense clouds, achieving an mIoU of 48.12%, while the Near-Infrared, green, and red (NGR) combination performed best in identifying cloud shadows with an mIoU of 23.32%. These findings highlight the critical role of radiometric correction and optimal channel selection in enhancing deep learning-based cloud detection. This study demonstrates the crucial role of radiometric correction, optimal channel selection, and the integration of additional synthetic indices in enhancing deep learning-based cloud detection performance, providing a foundation for the development of more refined cloud masking techniques in the future.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Solar Radiation and Photovoltaics · Calibration and Measurement Techniques · Atmospheric aerosols and clouds
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Earth and Environmental Sciences, Jeonbuk National University, Jeonju 54896, Republic of Korea
Korea Aerospace Research Institute, 169-84, Gwahak-ro, Yuseong, Daejeon 34133, Republic of Korea
Applied Plant Science, Chonnam National University, 77 Yongbong-ro, Buk-gu, Gwangju 61186, Republic of Korea
Funding
National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) · “Regional Innovation Strategy (RIS)” through the National Re-search Foundation of Korea(NRF) funded by the Ministry of Education (MOE)
References (34)
Show source-provided references
1. An assessment of thin cloud detection by applying bidirectional reflectance distribution function model-based background surface reflectance using Geostationary Ocean Color Imager (GOCI): A case study for South Korea10.1002/2017JD026707 ↗
2. A cloud masking algorithm for the XBAER aerosol retrieval using MERIS data10.1016/j.rse.2016.11.016 ↗
3. Greco, S., Infusino, M., De Donato, C., Coluzzi, R., Imbrenda, V., Lanfredi, M., Simoniello, T., and Scalercio, S. (2018). Late spring frost in Mediterranean beech forests: Extended crown dieback and short-term effects on moth communities. Forests, 9.10.3390/f9070388 ↗
4. Song, X., Yang, C., Wu, M., Zhao, C., Yang, G., Hoffmann, W., and Huang, W. (2017). Evaluation of Sentinel-2A satellite imagery for mapping cotton root rot. Remote Sens., 9.10.3390/rs9090906 ↗
5. Dörnhöfer, K., Göritz, A., Gege, P., Pflug, B., and Oppelt, N. (2016). Water constituents and water depth retrieval from Sentinel-2A—A first evaluation in an oligotrophic lake. Remote Sens., 8.10.3390/rs8110941 ↗
6. Object-based cloud and cloud shadow detection in Landsat imagery10.1016/j.rse.2011.10.028 ↗
7. Solar forecasting methods for renewable energy integration10.1016/j.pecs.2013.06.002 ↗
8. Validation of short and medium term operational solar radiation forecasts in the US10.1016/j.solener.2010.08.014 ↗
9. Semantic segmentation of clouds in satellite images based on U-Net++ architecture and attention mechanism10.1016/j.eswa.2022.118380 ↗
10. Main-Knorn, M., Pflug, B., Louis, J., Debaecker, V., Müller-Wilm, U., and Gascon, F. (2017). Sen2Cor for Sentinel-2. Image and Signal Processing for Remote Sensing XXIII, Presented at the Image and Signal Processing for Remote Sensing XXIII; SPIE.10.1117/12.2278218 ↗
11. Thin cloud detection over land using background surface reflectance based on the BRDF model applied to Geostationary Ocean Color Imager (GOCI) satellite data sets10.1016/j.rse.2019.111610 ↗
12. An improved method for detecting clear sky and cloudy radiances from AVHRR data10.1080/01431168808954841 ↗
13. Improvement and expansion of the Fmask algorithm: Cloud, cloud shadow, and snow detection for Landsats 4–7, 8, and Sentinel-2 images10.1016/j.rse.2014.12.014 ↗
14. Fmask 4.0: Improved cloud and cloud shadow detection in Landsats 4–8 and Sentinel-2 imagery10.1016/j.rse.2019.05.024 ↗
15. Ge, K., Liu, J., Wang, F., Chen, B., and Hu, Y. (2023). A Cloud Detection Method Based on Spectral and Gradient Features for SDGSAT-1 Multispectral Images. Remote Sens., 15.10.3390/rs15010024 ↗
16. CloudS2Mask: A novel deep learning approach for improved cloud and cloud shadow masking in Sentinel-2 imagery10.1016/j.rse.2024.114122 ↗
17. Zupanc, A., and Improving Cloud Detection with Machine Learning (2025, April 07). Sentin. Hub Blog., Available online: https://medium.com/sentinel-hub/improving-cloud-detection-with-machine-learning-c09dc5d7cf13.
18. A cloud detection algorithm for satellite imagery based on deep learning10.1016/j.rse.2019.03.039 ↗
19. Automated detection of cloud and cloud shadow in single-date Landsat imagery using neural networks and spatial post-processing10.3390/rs6064907 ↗
20. Multi-sensor cloud and cloud shadow segmentation with a convolutional neural network10.1016/j.rse.2019.05.022 ↗
21. A study on the cloud detection technique of heterogeneous sensors using Modified DeepLabV3+
22. Hughes, M.J., and Kennedy, R. (2019). High-quality cloud masking of Landsat 8 imagery using convolutional neural networks. Remote Sens., 11.10.3390/rs11212591 ↗
23. Morales, G., Ramirez, A., and Telles, J. (2019, January 29–31). End-to-end cloud segmentation in high-resolution multispectral satellite imagery using deep learning. Proceedings of the 2019 IEEE XXVI International Conference on Electronics, Electrical Engineering and Computing (INTERCON), Lima, Peru.10.1109/INTERCON.2019.8853549 ↗
24. Cloud detection in Landsat imagery of ice sheets using shadow matching technique and automatic normalized difference snow index threshold value decision10.1016/j.rse.2004.03.007 ↗
25. Multi-feature combined cloud and cloud shadow detection in GaoFen-1 wide field of view imagery10.1016/j.rse.2017.01.026 ↗
26. Kim, B., Lee, K., and Park, S. (2024). Burned-area mapping using post-fire PlanetScope images and a convolutional neural network. Remote Sens., 16.10.3390/rs16142629 ↗
27. Pešek, O., Segal-Rozenhaimer, M., and Karnieli, A. (2022). Using convolutional neural networks for cloud detection on VENμS images over multiple land-cover types. Remote Sens., 14.10.3390/rs14205210 ↗
28. Fabio, L., Piga, D., Umberto, M., and Safouane, E.G. (2024). BenchCloudVision: A benchmark analysis of deep learning approaches for cloud detection and segmentation in remote sensing imagery. arXiv.
29. Mohajerani, S., and Saeedi, P. (2018, January 29–31). CPNet: A context preserver convolutional neural network for detecting shadows in single RGB images. Proceedings of the 2018 IEEE 20th International Workshop on Multimedia Signal Processing (MMSP), Vancouver, BC, Canada.10.1109/MMSP.2018.8547080 ↗
30. Chen, L.-C., Zhu, Y., Papandreou, G., Schroff, F., and Adam, H. (2018). Encoder-decoder with atrous separable convolution for semantic image segmentation. arXiv.10.1007/978-3-030-01234-2_49 ↗
31. Mohajerani, S., and Saeedi, P. (2019). Cloud-Net: An end-to-end cloud detection algorithm for Landsat 8 imagery. arXiv.10.1109/IGARSS.2019.8898776 ↗
32. Chen, L.-C., Papandreou, G., Schroff, F., and Adam, H. (2017). Rethinking atrous convolution for semantic image segmentation. arXiv.
33. Milletari, F., Navab, N., and Ahmadi, S.-A. (2016, January 25–28). V-Net: Fully convolutional neural networks for volumetric medical image segmentation. Proceedings of the 2016 Fourth International Conference on 3D Vision (3DV), Stanford, CA, USA.10.1109/3DV.2016.79 ↗
34. Oh, H., Kim, B., and Kim, J. (2020, January 18–20). Deep learning-based cloud detection algorithm for KOMPSAT-3/3A imagery. Proceedings of the KSAS 2020 Fall Conference, Jeju, Republic of Korea.