UDC 535.8
CSCSTI 29.31
Russian Classification of Professions by Education 03.04.02
Russian Library and Bibliographic Classification 223
Russian Trade and Bibliographic Classification 6135
BISAK SCI053000 Physics / Optics & Light
The paper considers approaches to Allsky image segmentation for cloud detection. Color index-based methods, machine learning algorithms, and neural network models are compared. Color threshold methods, such as the cloud index, have shown high sensitivity with ease of implementation, but low versatility. Classic machine learning algorithms provide more flexible adaptation, but are limited in taking into account the spatial context. The results of neural network segmentation turned out to be less accurate, which is explained by the insufficient volume of the training sample. A conclusion is made about the prospects of combined solutions combining simple heuristics, machine learning, and deep neural networks to improve the accuracy and reliability of cloud analysis.
cloudiness, segmentation methods, machine learning algorithms
1. Galileyskiy V.P., Elizarov A.I., Kokarev D.V., Matvienko G.G., Morozov A.M. Programmno-tehnicheskaya sistema monitoringa i kontrolya optiko-fizicheskogo sostoyaniya atmosfery. Chast' 1. Panoramno-opticheskaya stanciya // Optika atmosfery i okeana. Fizika atmosfery: Sbornik dokladov XXVII Mezhdunarodnogo simpoziuma. [Elektronnyy resurs]. ISBN 978-5-94458-187-7. - Tomsk: Izd-vo IOA SO RAN, 2021. C. F5-F9.
2. Kalisch J. Estimation of the total cloud cover with high temporal resolution and parametrization of short-term fluctuations of sea surface insolation / Kalisch J., Macke A. // Meteorologische Zeitschrift – 2008. – T. 17 – № 5 – S. 603–611.
3. Long C.N. Retrieving Cloud Characteristics from Ground-Based Daytime Color All-Sky Images / Long C.N., Sabburg J.M., Calbó J., Pagès D. // Journal of Atmospheric and Oceanic Technology – 2006. – T. 23 – № 5 – S. 633–652.
4. Xie W. et al. SegCloud: A novel cloud image segmentation model using a deep convolutional neural network for ground-based all-sky-view camera observation //Atmospheric Measurement Techniques. – 2020. – T. 13. – №. 4. – S. 1953-1961.
5. Latry C., Panem C., Dejean P. Cloud detection with SVM technique //2007 IEEE International Geoscience and Remote Sensing Symposium. – IEEE, 2007. – S. 448-451.
6. Wei J. et al. Cloud detection for Landsat imagery by combining the random forest and superpixels extracted via energy-driven sampling segmentation approaches //Remote Sensing of Environment. – 2020. – T. 248. – S. 112005.
