结合辐射传输模拟与浅层神经网络的FY-3D MERSI影像云识别
Cloud detection for FY-3D MERSI Ⅱ images combine radiative transfer simulation and shallow neural network
- 2022年26卷第11期 页码：2136-2146
纸质出版日期： 2022-11-07
DOI： 10.11834/jrs.20210406
扫 描 看 全 文
浏览全部资源
扫码关注微信
摘要
本文结合辐射传输模型和机器学习提出了一种从FY-3D卫星MERSI Ⅱ传感器光学影像中识别云像元的方法CRMC（Combine Reflectance simulation and Machine learning for Cloud detection）。该方法通过设置变化的地物和大气内在光学特性IOPs（Inherent Optical Properties），达到考虑多种下垫面的二项反射特征和不同大气条件下气溶胶和云参数的目的。CRMC方法主要包含3个步骤：（1）通过聚类分析从MODIS二项反射参数产品中分离出11种典型下垫面地表反射参数；（2）将随机设置的气溶胶和云参数以及地表反射率参数（即IOPs）输入SBDART辐射传输模型，得到模拟的反射率值数据集，并以此训练浅层神经网络模型；（3）利用浅层神经网络模型逐像元预测云概率，并根据实际需要确定区分云像元和非云像元的云概率阈值。通过与CALIPSO垂直特性掩膜产品（VFM）逐像元对比验证发现，CRMC方法总正确率为79.6%，且在陆地和海面上分别为78.5%和81.2%。通过与MODIS云掩膜产品横向对比（MYD35）发现，当云阈值设定为0.2时，CRMC方法在陆地，主要是阔叶林、农田、城市和裸土等地表上的云识别效果较好，但在海面的云识别效果仍待进一步提高。
Abstract
Cloud pixel detection is a crucial pre-process in numerous remote sensing applications
such as the aerosol parameter retrieval
land use change
anomaly detection/classification
crop monitoring
and marine ecological survey. On the one hand
cloud pixel misidentification (mistaken as surface or aerosol) in optical images has substantial negative effects on the above-mentioned traditional applications
due to the significant influence of cloud layers on shortwave radiation. On the other hand
traditional threshold methods are largely constrained in applicability by the high spatiotemporal heterogeneity of aerosols and clouds as well as the diversification of satellite sensors and spectral channels. Therefore
a reliable cloud detection method with wide applicability is in demand
especially for studies over multiple surfaces (i.e.
land
ocean
and cryosphere). Thus
this study proposed an innovative method based on the radiative transfer simulation and machine learning
namely CRMC (Combine Reflectance simulation and Machine learning for Cloud detection)
to detect cloud pixels in optical images produced from the MERSI Ⅱ sensor onboard the FY-3D satellite. The biggest advantage of this method is its compatibility with different sensors
generating cloud pixel samples through the physical process simulation
and applying the machine learning technique to learn sample features
thereby excluding the influence of anthropogenic factors. Specifically
to address the mismatch between the MODIS cloud detection algorithm and MERSI-based aerosol inversion
the CRMC method sets different Inherent Optical Properties (IOPs) of surface and atmospheric objects
considering binomial reflection characteristics of underlying surfaces and various parameters of aerosols and clouds. In addition
the method outputs cloud probabilities in pixels and allows custom thresholds to control the strictness level of cloud pixel detection. The CRMC method mainly includes three steps: (1) Defining 11 typical underlying reflectance parameters from MODIS binomial reflection products using a cluster analysis approach; (2) Inputting the typical underlying reflectance and aerosol and cloud parameters with random inherent optical properties into the SBDART radiation transmission model to obtain a simulated reflectance dataset for training the shallow neural network; (3) Calculating the cloud probability of the target image with the trained shallow neural network and selecting a suitable threshold according to the actual need to complete the cloud detection. Compared with the CALIPSO Vertical Feature Mask (VFM)
results of the CRMC method show a maximum total accuracy of 79.6% (78.5% and 81.2% on land and sea
respectively). Under the condition of cloud probability threshold=0.2 (hit rates for cloud and cloud-free pixels are the same in these cases)
the CRMC outperforms MODIS cloud mask products (MYD35) over land
especially on broad-leaved forest
farmland
urban and bare soil. However
the accuracy of the CRMC over sea is lower than that of MYD35. In the view of the surface uniformity
the cloud detection over sea can enrich the brightness temperature information to optimize the corresponding performance. To sum up
while greatly improving the applicability to different optical sensors by not relying on special spectral range
the CRMC method can achieve a good cloud pixel identification effect for FY-3D MERSI Ⅱ images
with a similar hit rate compared with MODIS products. Also
the CRMC has certain anti-disturbance ability to haze.
关键词
云检测辐射传输模拟神经网络FY-3DMERSI II
Keywords
cloud detectionradiative transfer simulationneural networkFY-3DMERSI II
references
Ackerman S A, Strabala K I, Menzel W P, Frey R A, Moeller C C and Gumley L E. 1998. Discriminating clear sky from clouds with MODIS. Journal of Geophysical Research: Atmospheres, 103(D24): 32141-32157 [DOI: 10.1029/1998JD200032http://dx.doi.org/10.1029/1998JD200032]
Chen X D, Zhang X, Liu L Y and Wang X F. 2019. Enhanced multi-temporal cloud detection algorithm for optical remote sensing images. Journal of Remote Sensing, 23(2): 280-290
陈曦东, 张肖, 刘良云, 汪晓帆. 2019. 增强型多时相云检测. 遥感学报, 23(2): 280-290 [10.11834/jrs.20198017http://dx.doi.org/10.11834/jrs.20198017]
Gao B C, Goetz A F H and Wiscombe W J. 1993. Cirrus cloud detection from Airborne Imaging Spectrometer data using the 1.38 μm water vapor band. Geophysical Research Letters, 20(4): 301-304 [DOI: 10.1029/93GL00106http://dx.doi.org/10.1029/93GL00106]
Hughes M J and Hayes D J. 2014. Automated detection of cloud and cloud shadow in single-date landsat imagery using neural networks and spatial post-processing. Remote Sensing, 6(6): 4907-4926 [DOI: 10.3390/rs6064907http://dx.doi.org/10.3390/rs6064907]
Irish R R, Barker J L, Goward S N and Arvidson T. 2006. Characterization of the Landsat-7 ETM+ automated cloud-cover assessment (ACCA) algorithm. Photogrammetric Engineering and Remote Sensing, 72(10): 1179-1188 [DOI: 10.14358/PERS.72.10.1179http://dx.doi.org/10.14358/PERS.72.10.1179]
Jiao Z T, Li X W, Wang J D and Zhang H. 2011. Assessment of MODIS BRDF shape indicators. Journal of Remote Sensing, 15(3): 432-456
焦子锑, 李小文, 王锦地, 张虎. 2011. 评估MODIS的BRDF角度指数产品. 遥感学报, 15(3): 432-456 [DOI: 10.11834/jrs.20110073http://dx.doi.org/10.11834/jrs.20110073]
Jin S K, Zhang M, Ma Y Y, Gong W, Chen C, Yang L K, Hu X Q, Liu B M, Chen N, Du B and Shi Y F. 2021. Adapting the dark target algorithm to advanced MERSI sensor on the FengYun-3-D satellite: retrieval and validation of aerosol optical depth over land. IEEE Transactions on Geoscience and Remote Sensing, 59(10): 8781-8797 [DOI: 10.1109/TGRS.2020.3021021http://dx.doi.org/10.1109/TGRS.2020.3021021]
Kotchenova S Y, Vermote E F, Levy R and Lyapustin A. 2008. Radiative transfer codes for atmospheric correction and aerosol retrieval: intercomparison study. Applied Optics, 47(13): 2215-2226. [10.1364/AO.47.002215]
Li X H, Shen H F, Zhang L P, Zhang H Y, Yuan Q Q and Yang G. 2014. Recovering quantitative remote sensing products contaminated by thick clouds and shadows using multitemporal dictionary learning. IEEE Transactions on Geoscience and Remote Sensing, 52(11): 7086-7098 [DOI: 10.1109/TGRS.2014.2307354http://dx.doi.org/10.1109/TGRS.2014.2307354]
Li Z W, Shen H F, Cheng Q, Liu Y H, You S C and He Z Y. 2019. Deep learning based cloud detection for medium and high resolution remote sensing images of different sensors. ISPRS Journal of Photogrammetry and Remote Sensing, 150: 197-212[DOI: 10.1016/j.isprsjprs.2019.02.017http://dx.doi.org/10.1016/j.isprsjprs.2019.02.017]
Lin C H, Lin B Y, Lee K Y and Chen Y C. 2015. Radiometric normalization and cloud detection of optical satellite images using invariant pixels. ISPRS Journal of Photogrammetry and Remote Sensing, 106: 107-117 [DOI: 10.1016/j.isprsjprs.2015.05.003http://dx.doi.org/10.1016/j.isprsjprs.2015.05.003]
Liu C L and Wu B F. 2004. Application of cloud detection algorithm for the AVHRR data. Journal of Remote Sensing, 8(6):677-687
刘成林, 吴炳方. 2004. NOAA AVHRR云标识技术的应用分析. 遥感学报，8(6):677-687 [DOI: 10.11834/jrs.20040619http://dx.doi.org/10.11834/jrs.20040619]
McClatchey R A, Fenn R W, Selby J E A, Volz F E and Garing J S. 1971. Optical Properties of the Atmosphere. AFCRL-TR-71-0279, Environment Research Papers, No 354, L.G. Hancom Fiel Bedford, Mass. USA
Meng F, Yang X M, Zhou C H and Li Z. 2017. A sparse dictionary learning-based adaptive patch inpainting method for thick clouds removal from high-spatial resolution remote sensing imagery. Sensors (Basel), 17(9): 2130 [DOI: 10.3390/s17092130http://dx.doi.org/10.3390/s17092130]
Peng L K，Liu L C，Chen X H，Chen J，Cao X and Qiu Y A. 2021. Generalization ability of cloud detection network for satellite imagery based on DeepLabv3+. National Remote Sensing Bulletin，25（5）：1169-1186
彭龙康，刘励聪，陈学泓，陈晋，曹鑫，邱岳安. 2021.遥感影像云检测网络泛化性能研究：以DeepLabv3+为例. 遥感学报，25（5）：1169-1186 [DOI: 10.11834/jrs.20210061http://dx.doi.org/10.11834/jrs.20210061]
Ricchiazzi P, Yang S R, Gautier C and Sowle D. 1998. SBDART: a research and teaching software tool for plane-parallel radiative transfer in the Earth's atmosphere. Bulletin of the American Meteorological Society, 79(10): 2101-2114 [DOI: 10.1175/1520-0477(1998)079<2101:SARATS>2.0.CO;2http://dx.doi.org/10.1175/1520-0477(1998)079<2101:SARATS>2.0.CO;2]
Roujean J L, Leroy M and Deschamps P Y. 1992. A bidirectional reflectance model of the Earth’s surface for the correction of remote sensing data. Journal of Geophysical Research, 97(D18): 20455-20468 [DOI: 10.1029/92JD01411http://dx.doi.org/10.1029/92JD01411]
Segal-Rozenhaimer M, Li A L, Das K and Chirayath V. 2020. Cloud detection algorithm for multi-modal satellite imagery using convolutional neural-networks (CNN). Remote Sensing of Environment, 237: 111446 [DOI: 10.1016/j.rse.2019.111446http://dx.doi.org/10.1016/j.rse.2019.111446]
Stamnes K, Hamre B, Stamnes J J, Ryzhikov G, Biryulina M, Mahoney R, Hauss B and Sei A. 2011. Modeling of radiation transport in coupled atmosphere-snow-ice-ocean systems. Journal of Quantitative Spectroscopy and Radiative Transfer, 112(4): 714-726 [DOI: 10.1016/j.jqsrt.2010.06.006http://dx.doi.org/10.1016/j.jqsrt.2010.06.006]
Stowe L L, Davis P A and McClain E P. 1999. Scientific basis and initial evaluation of the CLAVR-1 global clear cloud classification algorithm for the advanced very high resolution radiometer. Journal of Atmospheric and Oceanic Technology, 16(6): 656-681 [DOI: 10.1175/1520-0426(1999)016<0656:SBAIEO>2.0.CO;2http://dx.doi.org/10.1175/1520-0426(1999)016<0656:SBAIEO>2.0.CO;2]
Sun L, Wei J, Wang J, Mi X T, Guo Y M, Lv Y, Yang Y K, Gan P, Zhou X Y, Jia C and Tian X P. 2016. A Universal Dynamic Threshold Cloud Detection Algorithm (UDTCDA) supported by a prior surface reflectance database. Journal of Geophysical Research: Atmospheres, 121(12): 7172-7196 [DOI: 10.1002/2015JD024722http://dx.doi.org/10.1002/2015JD024722]
Vermote E F, Tanre D, Deuze J L, Herman M and Morcette J J. 1997. Second simulation of the satellite signal in the solar spectrum, 6S: an overview. IEEE Transactions on Geoscience and Remote Sensing, 35(3): 675-686 [DOI: 10.1109/36.581987http://dx.doi.org/10.1109/36.581987]
Wei J, Huang W, Li Z Q, Sun L, Zhu X L, Yuan Q Q, Liu L and Cribb M. 2020. Cloud detection for Landsat imagery by combining the random forest and superpixels extracted via energy-driven sampling segmentation approaches. Remote Sensing of Environment, 248: 112005 [DOI: 10.1016/j.rse.2020.112005http://dx.doi.org/10.1016/j.rse.2020.112005]
Wei L S, Shang H Z, Husi L T, Ma R, Hu D H, Zhao K F, Si F Q and Shi J C. 2021. Cloud detection algorithm based on GF-5 DPC data. National Remote Sensing Bulletin, 25 (10)：2053-2066
伟乐斯，尚华哲，胡斯勒图，马润，胡大海，朝克夫，司福祺，施建成. 2021. GF-5 DPC数据的云检测方法研究. 遥感学报，25 (10)：2053-2066 [DOI: 10.11834/jrs.20210226http://dx.doi.org/10.11834/jrs.20210226]
Xu N, Niu X H, Hu X Q, Wang X H, Wu R H, Chen S S, Chen L, Sun L, Ding L, Yang Z D and Zhang P. 2018. Prelaunch calibration and radiometric performance of the advanced MERSI Ⅱ on FengYun-3D. IEEE Transactions on Geoscience and Remote Sensing, 56(8): 4866-4875 [DOI: 10.1109/TGRS.2018.2841827http://dx.doi.org/10.1109/TGRS.2018.2841827]
Yang Z D, Zhang P, Gu S Y, Hu X Q, Tang S H, Yang L K, Xu N, Zhen Z J, Wang L, Wu Q, Dou F L, Liu R X, Wu X, Zhu L, Zhang L Y, Wang S J, Sun Y Q and Bai W H. 2019. Capability of Fengyun-3D satellite in earth system observation. Journal of Meteorological Research, 33(6): 1113-1130 [DOI: 10.1007/s13351-019-9063-4http://dx.doi.org/10.1007/s13351-019-9063-4]
Yu C Y, Liang B, Bao C G, Li M, Hu Y Y, Lan D D, Xu Y, Lu H and Ma M H. 2013. Study on eutrophication status and trend in Bohai Sea. Marine Environmental Science, 32(2): 175-177
于春艳, 梁斌, 鲍晨光, 李冕, 胡莹莹, 兰冬东, 许妍, 洛昊, 马明辉. 2013. 渤海富营养化现状及趋势研究. 海洋环境科学, 32(2): 175-177 [DOI: CNKI:SUN:HYHJ.0.2013-02-004http://dx.doi.org/CNKI:SUN:HYHJ.0.2013-02-004]
Luo J C, Zhou C H and Yang Y. 2001. ANN Remote Sensing Classification Model and Its Integration Approach with Geo-knowledge. Journal of Remote Sensing, 5(2):122-129
骆剑承, 周成虎, 杨艳. 2001. 人工神经网络遥感影像分类模型及其与知识集成方法研究. 遥感学报, 5(2):122-129 [10.11834/jrs.20010210http://dx.doi.org/10.11834/jrs.20010210]
Zhang C, Li H F and Shen H F. 2020. Thin cloud correction for GF-5 AHSI visible images by combining statistical information with scattering model. Journal of Remote Sensing, 24(4): 368-378
张弛, 李慧芳, 沈焕锋. 2020. 联合统计信息与散射模型的GF-5 AHSI可见光影像薄云校正. 遥感学报, 24(4): 368-378 [10.11834/jrs.20209271http://dx.doi.org/10.11834/jrs.20209271]
Zhou H M, Yang W X and Lu X. 1995. Study on detecting cloud of NOAA satellite image. Journal of Remote Sensing, 10(2):137-142
周红妹, 杨星卫, 陆贤. 1995. NOAA气象卫星云检测方法的研究. 环境遥感, 10(2):137-142 [DOI: 10.11834/jrs.1995019http://dx.doi.org/10.11834/jrs.1995019]
Zhu Z and Woodcock C E. 2012. Object-based cloud and cloud shadow detection in Landsat imagery. Remote Sensing of Environment, 118: 83-94 [DOI: 10.1016/j.rse.2011.10.028http://dx.doi.org/10.1016/j.rse.2011.10.028]
文章被引用时，请邮件提醒。
提交
相关作者
暂无数据
相关机构
自然资源部 空间海洋遥感与应用重点实验室
自然资源部 国家卫星海洋应用中心
青岛海洋科学与技术试点国家实验室 区域海洋动力学与数值模拟功能实验室
中国海洋大学 三亚海洋研究院
中国海洋大学 信息科学与工程学部海洋技术学院
选择翻译语言
- 地址：北京市海淀区北四环西路19号 《遥感学报》编辑部 邮编：100190
- 联系电话：010-58887052 Email：nrsb@aircas.ac.cn
- 技术支持由北京北大方正电子有限公司提供 京ICP备20021838号-9
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies
0
回到顶部