Cloud detection of Sentinel-2 images for multiple backgrounds

doi:10.6046/zrzyyg.2022168

Abstract
Figures/Tables
References
Related Articles
Metrics

Download: PDF(7334 KB) HTML
Export: BibTeX | EndNote | Reference Manager | ProCite | RefWorks

Abstract

Cloud cover tends to hinder information extraction from remote sensing images during image processing. However, complex and changeable surface backgrounds make it difficult to effectively extract the differences in features between cloud targets and backgrounds. Although existing methods exhibit satisfactory cloud detection effects under most backgrounds, they show significant misclassification and omission in some environments, failing to maintain encouraging performance due to poor stability and insufficient generalization ability. Given this, this study proposed a cloud detection method for multiple backgrounds. Based on Sentinel-2A data, this study analyzed the differences in spectral characteristics between cloud targets and backgrounds to assist in the selection of samples for detection. Based on this, this study introduced more effective detection indices HOT and CDI. Finally, this study obtained a random forest-based cloud detection model through training. Then, from the perspective of the influence of backgrounds and cloud target types on detection accuracy, this study compared the obtained cloud detection model with the Fmask algorithm using images with different backgrounds. The comparison results show that the method proposed in this study increased the overall accuracy and F1 score by 2.2% and 2.9%, respectively, with the standard deviations of them reducing by 29.6% and 72.5%, respectively. These findings indicated that this method can significantly improve the stability of cloud detection in different environments while maintaining high detection accuracy. Therefore, this method is effective in cloud detection in multi-backgrounds.

Keywords Sentinel-2 cloud detection random forest Fmask HOT CDI multi-background environment

ZTFLH:

TP79

Issue Date: 19 September 2023

	Service

	E-mail this article
	E-mail Alert
	RSS
	Articles by authors

	Weichao WU
	Fawang YE

Cite this article:

Weichao WU,Fawang YE. Cloud detection of Sentinel-2 images for multiple backgrounds[J]. Remote Sensing for Natural Resources, 2023, 35(3): 124-133.

URL:

https://www.gtzyyg.com/EN/10.6046/zrzyyg.2022168 OR https://www.gtzyyg.com/EN/Y2023/V35/I3/124

Tab.1 Sentinel-2A satellite bands

Tab.2 Sample data

Fig.1 Flow chart of this study

Fig.2 Examples of opaque cloud and cirrus

Fig.3 TOA reflectance of typical cloud objects and land cover types

Fig.4 TOA reflectance of typical cloud objects and land cover types in B10

Tab.3 Validation data

Tab.4 Comparison of cloud detection accuracy

Fig.5 Results of cloud detection

Fig.6 Examples of commission errors in desert areas

Fig.7 Examples of commission errors in built-up areas

Fig.8 Examples of commission errors in ice/snow areas

Fig.9 Examples of commission errors of cirrus in high altitude

Fig.10 Examples of omission errors of thin cloud in low altitude

[1]	Rossow W B, Duenas E N. The international satellite cloud climatology project (ISCCP) web site:An online resource for research[J]. Bulletin of the American Meteorological Society, 2004, 85(2):167-172. doi: 10.1175/BAMS-85-2-173 url: https://journals.ametsoc.org/doi/10.1175/BAMS-85-2-173
[2]	Lu D, Weng Q. A survey of image classification methods and techniques for improving classification performance[J]. International Journal of Remote Sensing, 2007, 28(5):823-870. doi: 10.1080/01431160600746456 url: https://www.tandfonline.com/doi/full/10.1080/01431160600746456
[3]	Zhu X, Helmer E H. An automatic method for screening clouds and cloud shadows in optical satellite image time series in cloudy regions[J]. Remote Sensing of Environment, 2018, 214:135-153. doi: 10.1016/j.rse.2018.05.024 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425718302530
[4]	彭龙康, 刘励聪, 陈学泓, 等. 遥感影像云检测网络泛化性能研究:以DeepLabv3+为例[J]. 遥感学报, 2021, 25(5):1169-1186.
[4]	Peng L K, Liu L C, Chen X H, et al. Generalization ability of cloud detection network for satellite imagery based on DeepLabv3+[J]. National Remote Sensing Bulletin, 2021, 25(5):1169-1186. doi: 10.11834/jrs.20210061 url: http://www.ygxb.ac.cn/zh/article/doi/10.11834/jrs.20210061/
[5]	Coluzzi R, Imbrenda V, Lanfredi M, et al. A first assessment of the Sentinel-2 level 1-C cloud mask product to support informed surface analyses[J]. Remote Sensing of Environment, 2018, 217:426-443. doi: 10.1016/j.rse.2018.08.009 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425718303742
[6]	Hagolle O, Huc M, Pascual D V, et al. A multi-temporal method for cloud detection,applied to FORMOSAT-2,VENμS,LANDSAT and SENTINEL-2 images[J]. Remote Sensing of Environment, 2010, 114(8):1747-1755. doi: 10.1016/j.rse.2010.03.002 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425710000908
[7]	Main-Knorn M, Pflug B, Louis J, et al. Sen2Cor for Sentinel-2[C]// Image and Signal Processing for Remote Sensing XXIII.International Society for Optics and Photonics, 2017, 10427:37-48.
[8]	Qiu S, Zhu Z, He B. Fmask 4.0:Improved cloud and cloud shadow detection in Landsats4-8 and Sentinel-2 imagery[J]. Remote Sensing of Environment, 2019, 231:111205. doi: 10.1016/j.rse.2019.05.024 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425719302172
[9]	Fu H, Shen Y, Liu J, et al. Cloud detection for FY meteorology satellite based on ensemble thresholds and random forests approach[J]. Remote Sensing, 2019, 11(1):44. doi: 10.3390/rs11010044 url: http://www.mdpi.com/2072-4292/11/1/44
[10]	栗旭升, 刘玉峰, 陈冬花, 等. 结合图像特征的支持向量机高分一号云检测[J]. 国土资源遥感, 2020, 32(3):55-62.doi:10.6046/gtzyyg.2020.03.08. doi: 10.6046/gtzyyg.2020.03.08
[10]	Li X S, Liu Y F, Chen D H, et al. Cloud detection based on support vector machine with image features for GF-1 data[J]. Remote Sensing for Land and Resource, 2020, 32(3):55-62.doi:10.6046/gtzyyg.2020.03.08. doi: 10.6046/gtzyyg.2020.03.08
[11]	Ishida H, Oishi Y, Morita K, et al. Development of a support vector machine based cloud detection method for MODIS with the adjustability to various conditions[J]. Remote Sensing of Environment, 2018, 205:390-407. doi: 10.1016/j.rse.2017.11.003 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425717305138
[12]	Cilli R, Monaco A, Amoroso N, et al. Machine learning for cloud detection of globally distributed Sentinel-2 images[J]. Remote Sensing, 2020, 12(15):2355. doi: 10.3390/rs12152355 url: https://www.mdpi.com/2072-4292/12/15/2355
[13]	王明, 刘正佳, 陈元琰. 基于Sentinel-2波段/产品的图像云检测效果对比研究[J]. 遥感技术与应用, 2020, 35(5):1167-1177.
[13]	Wang M, Liu Z J, Chen Y Y. Comparsions of image cloud detection effect based on Sentinel-2 bands/products[J]. Remote Sensing Technology and Application, 2020, 35(5):1167-1177.
[14]	余长慧, 于海威, 张文, 等. 神经网络支持下的Sentinel-2卫星影像自动云检测[J]. 测绘通报, 2019(8):39-43. doi: 10.13474/j.cnki.11-2246.2019.0248
[14]	Yu C H, Yu H W, Zhang W, et al. Automatic cloud detection of Sentinel-2 satellite images based on neural network[J]. Bulletin of Surveying and Mapping, 2019(8):39-43. doi: 10.13474/j.cnki.11-2246.2019.0248
[15]	Hollstein A, Segl K, Guanter L, et al. Ready-to-use methods for the detection of clouds,cirrus,snow,shadow,water and clear sky pixels in Sentinel-2 MSI images[J]. Remote Sensing, 2016, 8(8):666. doi: 10.3390/rs8080666 url: http://www.mdpi.com/2072-4292/8/8/666
[16]	Zhang Y, Guindon B, Cihlar J. An image transform to characterize and compensate for spatial variations in thin cloud contamination of Landsat images[J]. Remote Sensing of Environment, 2002, 82(2-3):173-187. doi: 10.1016/S0034-4257(02)00034-2 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425702000342
[17]	姜杰, 查勇, 袁杰, 等. 遥感技术在灰霾监测中的应用综述[J]. 环境监测管理与技术, 2011, 23(2):15-18.
[17]	Jiang J, Zha Y, Yuan J, et al. Application review of remote sensing technology in haze monitoring[J]. The Administration and Technique of Environmental Monitoring, 2011, 23(2):15-18.
[18]	Breiman L. Random forests[J]. Machine Learning, 2001, 45(1):5-32. doi: 10.1023/A:1010933404324 url: http://link.springer.com/10.1023/A:1010933404324
[19]	Frantz D, Haß E, Uhl A, et al. Improvement of the Fmask algorithm for Sentinel-2 images:Separating clouds from bright surfaces based on parallax effects[J]. Remote Sensing of Environment, 2018, 215:471-481. doi: 10.1016/j.rse.2018.04.046 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425718302037
[20]	Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn:Machine learning in Python[J]. Journal of Machine Learning Research, 2011, 12:2825-2830.
[21]	Oshiro T M, Perez P S, Baranauskas J A. How many trees in a random forest?[C]// International Workshop on Machine Learning and Data Mining in Pattern Recognition.Springer,Berlin,Heidelberg, 2012:154-168.
[22]	Wei J, Huang W, Li Z, et al. Cloud detection for Landsat imagery by combining the random forest and superpixels extracted via energy-driven sampling segmentation approaches[J]. Remote Sensing of Environment, 2020, 248:112005. doi: 10.1016/j.rse.2020.112005 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425720303758
[23]	Zhu X, Helmer E H. An automatic method for screening clouds and cloud shadows in optical satellite image time series in cloudy regions[J]. Remote Sensing of Environment, 2018, 214:135-153. doi: 10.1016/j.rse.2018.05.024 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425718302530
[24]	Joachim L, Storch T. Cloud detection for night-time panchromatic visible and near-infrared satellite imagery[J]. ISPRS Annals of Photogrammetry,Remote Sensing and Spatial Information Sciences, 2020, 2:853-860.
[25]	Qiu S, He B, Zhu Z, et al. Improving Fmask cloud and cloud shadow detection in mountainous area for Landsats4-8 images[J]. Remote Sensing of Environment, 2017, 199:107-119. doi: 10.1016/j.rse.2017.07.002 url: https://linkinghub.elsevier.com/retrieve/pii/S0034425717303073

[1]	XI Lei, SHU Qingtai, SUN Yang, HUANG Jinjun, SONG Hanyue. Optimizing an ICESat2-based remote sensing estimation model for the leaf area index of mountain forests in southwestern China[J]. Remote Sensing for Natural Resources, 2023, 35(3): 160-169.
[2]	PARIHA Helili, ZAN Mei. Spatio-temporal changes and influencing factors of ecological environments in oasis cities of arid regions[J]. Remote Sensing for Natural Resources, 2023, 35(3): 201-211.
[3]	LIANG Jintao, CHEN Chao, ZHANG Zili, LIU Zhisong. A random forest-based method integrating indices and principal components for classifying remote sensing images[J]. Remote Sensing for Natural Resources, 2023, 35(3): 35-42.
[4]	CHEN Kai, WANG Chun, DAI Wen, SHENG Yehua, LIU Aili, TANG Guoan. Factors influencing the terrain modeling accuracy of UAV photogrammetry based on Monte Carlo tests of control points[J]. Remote Sensing for Natural Resources, 2023, 35(3): 107-115.
[5]	HOU Yingzhuo, JI Ling, XING Qianguo, SHENG Dezhi. Satellite remote sensing-assisted comparative monitoring of dynamic characteristics of macroalgae aquaculture in Weihai City, Shandong Province, China[J]. Remote Sensing for Natural Resources, 2023, 35(2): 34-41.
[6]	WANG Jue, GUO Zhen, ZHANG Zhiwei, XU Wenxue, XU Hao. Construction of 3D landscape indices based on tilt photogrammetry: A case study of Tianheng Island in Shandong Province[J]. Remote Sensing for Natural Resources, 2023, 35(2): 112-121.
[7]	ZHANG Tian, ZHOU Zhongfa, WANG Lingyu, ZHAO Xin, ZHANG Wenhui, ZHANG Shu, WANG Yu. A method for soil roughness measurement based on UAV point cloud data[J]. Remote Sensing for Natural Resources, 2023, 35(1): 115-122.
[8]	WU Yuxin, WANG Juanle, HAN Baomin, YAN Xinrong. Forest classification for Motuo County: A method based on spatio-temporal-spectral characteristics[J]. Remote Sensing for Natural Resources, 2023, 35(1): 180-188.
[9]	TIAN Chen, ZHANG Jinlong, JIN Yirong, DONG Shiyuan, WANG Bin, ZHANG Naixiang. A remote sensing classification method for cyanobacteria using Bayesian optimization algorithm[J]. Remote Sensing for Natural Resources, 2023, 35(1): 49-56.
[10]	ZHANG Hao, GAO Xiaohong, SHI Feifei, LI Runxiang. Sentinel-2 MSI and Sentinel-1 SAR based information extraction of abandoned land in the western Loess Plateau:A case study of Minhe County in Qinghai[J]. Remote Sensing for Natural Resources, 2022, 34(4): 144-154.
[11]	YAN Junjie, GUO Xuexing, QU Jianhua, HAN Min. An FY-4A/AGRI cloud detection model based on the naive Bayes algorithm[J]. Remote Sensing for Natural Resources, 2022, 34(3): 33-42.
[12]	WANG Chunxia, ZHANG Jun, LI Yixu, PHOUMILAY. The construction and verification of a water index in the complex environment based on GF-2 images[J]. Remote Sensing for Natural Resources, 2022, 34(3): 50-58.
[13]	DENG Jingwen, TIAN Yichao, ZHANG Qiang, TAO Jin, ZHANG Yali, HUANG Shengguang. Application of airborne LiDAR in the estimation of the mean height of mangrove stand[J]. Remote Sensing for Natural Resources, 2022, 34(3): 129-137.
[14]	LIU Guangjin, WANG Guanghui, BI Weihua, LIU Huijie, YANG Huachao. Cloud detection algorithm of remote sensing image based on DenseNet and attention mechanism[J]. Remote Sensing for Natural Resources, 2022, 34(2): 88-96.
[15]	WANG Xuejie, SHI Guoping, ZHOU Ziqin, ZHEN Yang. Revision of solar radiation product ERA5 based on random forest algorithm[J]. Remote Sensing for Natural Resources, 2022, 34(2): 105-111.

Viewed

Full text

Abstract

Cited

Shared

Discussed