Detection of severe convective cloud over sea surface from geostationary meteorological satellite images based on deep learning

Yiqin ZHENG; Xiaofeng YANG; Ziwei LI

doi:10.11834/jrs.20208209

Remote Sensing Applications | Views : 0 下载量: 140 CSCD: 6

PDF
Export
Share
Collection
Album

Detection of severe convective cloud over sea surface from geostationary meteorological satellite images based on deep learning
Vol. 24, Issue 1, Pages: 97-106(2020)
DOI： 10.11834/jrs.20208209

扫描看全文

Quote
郑益勤,杨晓峰,李紫薇.2020.深度学习模型识别静止卫星图像海上强对流云团.遥感学报,24(1): 97-106

ZHENG Yiqin,YANG Xiaofeng,LI Ziwei. 2020. Detection of severe convective cloud over sea surface from geostationary meteorological satellite images based on deep learning. Journal of Remote Sensing(Chinese). 24(1): 97-106
郑益勤,杨晓峰,李紫薇.2020.深度学习模型识别静止卫星图像海上强对流云团.遥感学报,24(1): 97-106 DOI： 10.11834/jrs.20208209.

ZHENG Yiqin,YANG Xiaofeng,LI Ziwei. 2020. Detection of severe convective cloud over sea surface from geostationary meteorological satellite images based on deep learning. Journal of Remote Sensing(Chinese). 24(1): 97-106 DOI： 10.11834/jrs.20208209.

摘要

强对流天气破坏力强，对海上航行和海洋开发都有着很大的影响，但由于其生命史短，且海上气象测站少，因此难以对其进行快速准确的监测。而静止气象卫星的覆盖范围广、时间分辨率高，因此成为了监测强对流天气的重要手段。本文提出了一种利用Himawari-8卫星影像，基于深度信念网络(DBN)进行强对流云团自动识别的方法。该方法分别提取每张图像的光谱特征TBB,13,、TBB,08,-TBB,13,和TBB,13,-TBB,15,，以及基于光谱特征TBB,08,-TBB,13,的纹理特征能量Energy和对比度Contrast，再参考CloudSat卫星的云分类产品自动构建样本集，利用此样本集训练DBN模型，以确定模型的参数和结构。使用训练完成的DBN模型进行强对流云团识别，并对识别结果进行后处理。通过典型案例分析和精度评定发现，新方法的临界成功指数CSI为71.28%，检测概率POD为84.83%，虚警率FAR为18.31%。结果表明，该方法可以有效识别处于初生到消散不同阶段的强对流云团，并在一定程度上去除检测结果中多余的卷云。与单波段阈值法、多波段阈值法和支持向量机这3种方法相比，文中提出的方法能够提高强对流云团的识别精度。

Abstract

Disastrous weather caused by Severe Convective Clouds (SCC), such as short-term heavy rainfall, hail, thunderstorms, squall line and tornado, has a serious impact on the life and production over sea surface. Therefore it is very essential to monitor SCC duly and accurately. However, the traditional methods are difficult to meet the requirements due to the short life cycle of severe convective weather and the lack of meteorological stations over sea surface. The geostationary meteorological satellite images have wide coverage area and high temporal resolution, so it has become an important means to monitor severe convective weather. With the rapid development of deep learning, it has been applied in the field of remote sensing image recognition.,This paper proposes a method for automatic identification of SCC from geostationary meteorological satellite images based on Deep Belief Networks (DBN). First, in order to identify the characteristic parameters that can be used for recognizing severe convective clouds, the severe convective clouds are analyzed from spectral features and texture features. Second, building automatic recognition algorithm based on deep belief networks which consists of multiple Restricted Boltzmann Machines (RBM) and a softmax classifier. And it is divided into two stages: unsupervised pre training and supervised fine tuning. This method is summarized in the following four steps: (1)data preprocessing: images splicing and region cutting; (2)extracting features and constructing sample sets: extracting the spectral features: TBB13, TBB08-TBB13 and TBB13-TBB15. And extracting the texture features: Energy and Contrast, which extracted based on spectral feature TBB08-TBB13. And then constructing the sample sets automatically referring to CloudSat satellite cloud classification products;(3)training DBN model: determining the structure and parameters of the model, including parameters of RBM and depth of DBN;(4)recognizing SCC using DBN model, and doing the postprocessing: identifying the severe convective clouds using DBN model, and processing the recognition results by category merging, closing operation and edge detection.,The Himawari-8 satellite image data and CloudSat cloud classification products from March to May in 2017 were used in the experiment. The research area is 70°E—150°E, 0°N—55°N. After training, the structure of DBN model was setting to 245-140-140-140-135-135-135-9. The accuracy of the DBN model is evaluated with the modified test sample, the Critical Success Index (CSI) is 71.28%, the Probability of Detection (POD) is 84.83%, and the False Alarm Ratio (FAR) is 18.31%. Compared with single band threshold method, multi band threshold method and SVM, the method proposed in this paper can effectively improve the recognition accuracy.,The experiment results show that all kinds of severe convective clouds in different phases from initiation to dissipation can be effectively identified. Consequently, it has the advantages of finding severe convective weather in advance. The majority of the cirrus can be removed, but the results of recognition still contain some cirrus spissatus, and the recognition of the cloud edge is not accurate enough, which will be two directions for future research.

关键词

遥感强对流云团深度信念网络静止气象卫星光谱特征纹理特征

Keywords

remote sensingsevere convective cloudsdeep belief networksgeostationary meteorological satellitespectral featuretexture feature

references

Ackerman S A. 1996. Global satellite observations of negative brightness temperature differences between 11 and 6.7 µm. Journal of the Atmospheric Sciences, 53(19): 2803-2812 [DOI: 10.1175/1520-0469(1996)053<2803:GSOONB>2.0.CO;2http://dx.doi.org/10.1175/1520-0469(1996)053<2803:GSOONB>2.0.CO;2]

Bedka K, Brunner J, Dworak R, Feltz W, Otkin J and Greenwald T. 2010. Objective satellite-based detection of overshooting tops using infrared window channel brightness temperature gradients. Journal of Applied Meteorology and Climatology, 49(2): 181-202 [DOI: 10.1175/2009JAMC2286.1http://dx.doi.org/10.1175/2009JAMC2286.1]

Chen Y S, Lin Z H, Zhao X, Wang G and Gu Y F. 2014. Deep learning-based classification of hyperspectral data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 7(6): 2094-2107 [DOI: 10.1109/JSTARS.2014.2329330http://dx.doi.org/10.1109/JSTARS.2014.2329330]

Fu R, Del Genio A D and Rossow W B. 1990. Behavior of deep convective clouds in the tropical Pacific deduced from ISCCP radiances. Journal of Climate, 3(10): 1129-1152 [DOI: 10.1175/1520-0442(1990)003<1129:BODCCI>2.0.CO;2http://dx.doi.org/10.1175/1520-0442(1990)003<1129:BODCCI>2.0.CO;2]

Hall T J and Vonder Haar T H. 1999. The diurnal cycle of west Pacific deep convection and its relation to the spatial and temporal variation of tropical MCSs. Journal of the Atmospheric Sciences, 56(19): 3401-3415 [DOI: 10.1175/1520-0469(1999)056<3401:TDCOWP>2.0.CO;2http://dx.doi.org/10.1175/1520-0469(1999)056<3401:TDCOWP>2.0.CO;2]

Hinton G E. 2002. Training products of experts by minimizing contrastive divergence. Neural Computation, 14(8): 1771-1800 [DOI: 10.1162/089976602760128018http://dx.doi.org/10.1162/089976602760128018]

Hinton G E, Osindero S and Teh Y W. 2006. A fast learning algorithm for deep belief nets. Neural Computation, 18(7): 1527-1554 [DOI: 10.1162/neco.2006.18.7.1527http://dx.doi.org/10.1162/neco.2006.18.7.1527]

Inoue T. 1987. A cloud type classification with NOAA 7 split-window measurements. Journal of Geophysical Research: Atmospheres, 92(D4): 3991-4000 [DOI: 10.1029/JD092iD04p03991http://dx.doi.org/10.1029/JD092iD04p03991]

Liu J W. 2014. Impact of underlaying surface on thunderstorm in China. Meteorological, Hydrological and Marine Instruments, 31(3): 125-128

刘佳伟. 2014. 中国地区下垫面对雷暴的影响研究进展. 气象水文海洋仪器, 31(3): 125-128）[DOI: 10.3969/j.issn.1006-009X.2014.03.032http://dx.doi.org/10.3969/j.issn.1006-009X.2014.03.032]

Ma Z S, Liu Q J, Qin Y Y and Chen Y. 2008. Introductions to a new type cloud detecting satellite-CloudSat. Meteorological Monthly, 34(8): 104-111

马占山, 刘奇俊, 秦琰琰, 谌芸. 2008. 云探测卫星CloudSat. 气象, 34(8): 104-111） [DOI: 10.7519/j.issn.1000-0526.2008.08.016http://dx.doi.org/10.7519/j.issn.1000-0526.2008.08.016]

Mecikalski J R and Bedka K M. 2006. Forecasting convective initiation by monitoring the evolution of moving cumulus in daytime GOES imagery. Monthly Weather Review, 134(1): 49-78 [DOI: 10.1175/MWR3062.1http://dx.doi.org/10.1175/MWR3062.1]

Mnih V and Hinton G E. 2010. Learning to detect roads in high-resolution aerial images//Proceedings of the 11th European Conference on Computer Vision. Heraklion, Crete, Greece: Springer: 210-223 [DOI: 10.1007/978-3-642-15567-3_16http://dx.doi.org/10.1007/978-3-642-15567-3_16]

Okuyama A, Andou A, Date K, Hoasaka K, Mori N, Murata H, Tabata T, Takahashi M, Yoshino R and Bessho K. 2015. Preliminary validation of Himawari-8/AHI navigation and calibration//Proceedings Volume 9607, Earth Observing Systems XX. San Diego, California, United States: SPIE: 96072E [DOI: 10.1117/12.2188978http://dx.doi.org/10.1117/12.2188978]

Szejwach G. 1982. Determination of semi-transparent cirrus cloud temperature from infrared radiances: application to METEOSAT. Journal of Applied Meteorology, 21(3): 384-393 [DOI: 10.1175/1520-0450(1982)021<0384:DOSTCC>2.0.CO;2http://dx.doi.org/10.1175/1520-0450(1982)021<0384:DOSTCC>2.0.CO;2]

Tokuno M. 1996. Operational system for extracting cloud motion and water vapor motion winds from GMS-5 image data//Proceedings of the 3rd International Winds Workshop. Ascona, Switzerland: EUMETSAT

Vaduva C, Gavat I, Datcu M. 2012. Deep learning in very high resolution remote sensing image information mining communication concept//Proceedings of the 20th European Signal Processing Conference. Bucharest, Romania: IEEE: 2506-2510

Vila D A, Machado L A T, Laurent H and Velasco I. 2008. Forecast and tracking the evolution of cloud clusters (ForTraCC) using satellite infrared imagery: methodology and validation. Weather and Forecasting, 23(2): 233-245 [DOI: 10.1175/2007WAF2006121.1http://dx.doi.org/10.1175/2007WAF2006121.1]

Welch R M, Sengupta S K, Goroch A K, Rabindra P, Rangaraj N and Navar M S. 1992. Polar cloud and surface classification using AVHRR imagery: an intercomparison of methods. Journal of Applied Meteorology, 31(5): 405-420 [DOI: 10.1175/1520-0450(1992)031<0405:PCASCU>2.0.CO;2http://dx.doi.org/10.1175/1520-0450(1992)031<0405:PCASCU>2.0.CO;2]

Xiao W P. 2015. Japanese first next-generation weather satellite Himawari-8 enters geostationary orbit. Space International, (2): 87-91

肖武平. 2015. 日本首颗新一代气象卫星向日葵-8入轨. 国际太空, (2): 87-91

Yan T Y and Wang S. 2015. The GHSOM network cloud classification model of stationary satellite infrared cloud Images at night. Journal of Jiangxi Normal University (Natural Science), 39(4): 383-388, 410

闫廷亚, 王杉. 2015. 夜间静止卫星红外云图的GHSOM网络云分类模型. 江西师范大学学报(自然科学版), 39(4): 383-388, 410） [DOI: 10.16357/j.cnki.issn1000-5862.2015.04.10http://dx.doi.org/10.16357/j.cnki.issn1000-5862.2015.04.10]

Zhang Y L, Liu G X, Cao D and Wu T W. 2007. Basic operators of mathematical morphology and application in image preprocessing. Science Technology and Engineering, 7(3): 356-359

张艳玲, 刘桂雄, 曹东, 吴庭万. 2007. 数学形态学的基本算法及在图像预处理中应用. 科学技术与工程, 7(3): 356-359） [DOI: 10.3969/j.issn.1671-1815.2007.03.021http://dx.doi.org/10.3969/j.issn.1671-1815.2007.03.021]

Zinner T, Mannstein H and Tafferner A. 2008. Cb-TRAM: tracking and monitoring severe convection from onset over rapid development to mature phase using multi-channel Meteosat-8 SEVIRI data. Meteorology and Atmospheric Physics, 101(3/4): 191-210 [DOI: 10.1007/s00703-008-0290-yhttp://dx.doi.org/10.1007/s00703-008-0290-y]

Alert me when the article has been cited

提交

Temporal sequence method for fire spot detection using Himawari-8 geostationary meteorological satellite

Spatio-temporal characteristics of CO₂ in global land area from 2010 to 2020 based on GOSAT satellite data

Satellite-marine synergistic extraction and analysis of reference temperature of thermal plume from nuclear power by taking Ningde as an example

Transfer learning method for landslide extraction from GF-1 images after the Wenchuan earthquake

Tropospheric NO₂ changes in global typical regions during the COVID-19 pandemic from GF-5 EMI observations