宽度学习系统的SAR影像海面强降雨智能检测研究
Intelligent detection of rain cells with SAR imagery based on broad learning system
- 2023年27卷第7期 页码：1605-1614
纸质出版日期： 2023-07-07
DOI： 10.11834/jrs.20221800
扫 描 看 全 文
浏览全部资源
扫码关注微信
摘要
海洋降雨对全球大气循环和局地气候均有重要影响，从遥感影像中监测降雨团对于海洋天气预报具有重要意义。合成孔径雷达以高空间分辨率进行大范围观测的能力使其成为10—30 km尺度大小的强降雨团的有效观测手段之一。针对Sentinel-1波模式获取的9种海面现象的SAR影像组成的数据集，本文使用融合特征的宽度学习系统BLS（Broad Learning System）进行了海面强降雨团的智能检测研究。结果表明强降雨团的检测精度为98.51%，召回率为95.24%，该结果与ResNet50预训练模型的结果相当，但是同等计算条件下后者的模型训练时间却是前者的20倍。此外，与传统深度学习网络相比，BLS的结构是灵活的，即可以通过增加节点或增加数据集来优化、更新模型。对于BLS的节点增量学习功能，本文实验证实其可以在无需重训练整个模型的前提下更新模型。针对训练数据集增广导致的模型更新任务，本文综合利用增量学习方案和重训练方案的优点提出了模型的混合更新方案，新方案既能保证模型的高精度又能显著降低模型更新所需时间。
Abstract
Ocean rainfall has an important impact on the global atmospheric cycle and local climate. Monitoring rain cells from remote sensing images is vital for ocean weather prediction. The ability of Synthetic Aperture Radar (SAR) to probe with a wide swath and high spatial resolution makes it an effective observation approach for rain cells with a scale of 10—30 km. This study uses the fusion-feature-based Broad Learning System (BLS) to detect the rain cells. The SAR images dataset composed of nine sea surface phenomena obtained by Sentinel-1 wave mode is also used. Results show that the detection accuracy is 98.51%
and the recall rate is 95.24%. These values are equivalent to those of the ResNet50 pretrained model. However
the training time of ResNet50 is 20 times that of BLS under the same calculation conditions. Compared with the structure of a deep learning network
that of BLS is flexible. That is
the model can be optimized and updated by adding nodes or input data. The experiments show that the node incremental learning of BLS can update the model without retraining the whole model. Following the advantages of the incremental learning and retraining schemes
this study proposed a hybrid model-updating scheme for the model-updating task caused by the expansion of the training dataset. This new scheme can ensure the high accuracy of the model and significantly reduce the time cost for model updating.
关键词
人工智能检测海面强降雨检测宽度学习系统合成孔径雷达模型更新
Keywords
artificial intelligence detectionrain cells detectionbroad learning systemsynthetic aperture radarmodel updating
references
Alpers W, Cheng C M, Shao Y and Yang L M. 2007. Study of rain events over the South China Sea by synergistic use of multi-sensor satellite and ground-based meteorological data. Photogrammetric Engineering and Remote Sensing, 73(3): 267-278 [DOI: 10.14358/pers.73.3.267http://dx.doi.org/10.14358/pers.73.3.267]
Alpers W and Melsheimer C. 2004. Rainfall//Jackson C R and Apel J R, eds. Synthetic Aperture Radar Marine User’s Manual. Washington: National Oceanic and Atmospheric Administration: 355-371
Atlas D. 1994. Footprints of storms on the sea: a view from spaceborne synthetic aperture radar. Journal of Geophysical Research: Oceans, 99(C4): 7961-7969 [DOI: 10.1029/94jc00250http://dx.doi.org/10.1029/94jc00250]
Bergstra J, Yamins D and Cox D D. 2013. Making a science of model search: hyperparameter optimization in hundreds of dimensions for vision architectures//Proceedings of the 30th International Conference on International Conference on Machine Learning. Atlanta: JMLR.org: I-115-I-123
Chan P W, Cheng C M and Alpers W. 2010. Study of wind fields associated with subtropical squall lines using Envisat Synthetic Aperture Radar images and ground-based weather radar data. International Journal of Remote Sensing, 31(17/18): 4897-4914 [DOI: 10.1080/01431161.2010.485144http://dx.doi.org/10.1080/01431161.2010.485144]
Chen C L P and Liu Z L. 2018. Broad learning system: an effective and efficient incremental learning system without the need for deep architecture. IEEE Transactions on Neural Networks and Learning Systems, 29(1): 10-24 [DOI: 10.1109/tnnls.2017.2716952http://dx.doi.org/10.1109/tnnls.2017.2716952]
Contreras R F and Plant W J. 2006. Surface effect of rain on microwave backscatter from the ocean: measurements and modeling. Journal of Geophysical Research: Oceans, 111(C8): C08019 [DOI: 10.1029/2005jc003356http://dx.doi.org/10.1029/2005jc003356]
Davis C, Brown B and Bullock R. 2006. Object-based verification of precipitation forecasts. Part I: methodology and application to mesoscale rain areas. Monthly Weather Review, 134(7): 1772-1784 [DOI: 10.1175/mwr3145.1http://dx.doi.org/10.1175/mwr3145.1]
Gan X L, Hunag W G, Yang J S, Lou X L, Shi A Q, Zhang X L and Sun X M. 2007. The study of fine structure of a mesoscale thunderstorm from SAR image. Remote Sensing Technology and Application, 22(2): 246-250
甘锡林, 黄韦艮, 杨劲松, 楼琇林, 史爱琴, 张小丽, 孙向明. 2007. 利用合成孔径雷达研究中尺度雷暴的细结构. 遥感技术与应用, 22(2): 246-250 [DOI: 10.3969/j.issn.1004-0323.2007.02.025http://dx.doi.org/10.3969/j.issn.1004-0323.2007.02.025]
Guo R F and Liu Y B. 2018. Strategy and method for satellite precipitation data evaluation: an overview. Remote Sensing Technology and Application, 33(6): 983-993
郭瑞芳, 刘元波. 2018. 遥感降水数据精度检验策略及检验方法综述. 遥感技术与应用, 33(6): 983-993 [DOI: 10.11873/j.issn.1004-0323.2018.6.0983http://dx.doi.org/10.11873/j.issn.1004-0323.2018.6.0983]
He K M, Zhang X Y, Ren S Q and Sun J. 2016. Deep residual learning for image recognition//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas: IEEE: 770-778 [DOI: 10.1109/CVPR.2016.90http://dx.doi.org/10.1109/CVPR.2016.90]
Houze R A. 1997. Stratiform precipitation in regions of convection: a meteorological paradox?. Bulletin of the American Meteorological Society, 78(10): 2179-2196 [DOI: 10.1175/1520-0477(1997)078<2179:Spiroc>2.0.Co;2http://dx.doi.org/10.1175/1520-0477(1997)078<2179:Spiroc>2.0.Co;2]
Kuok S C and Yuen K V. 2020a. Broad learning for nonparametric spatial modeling with application to seismic attenuation. Computer-Aided Civil and Infrastructure Engineering, 35(3): 203-218 [DOI: 10.1111/mice.12494http://dx.doi.org/10.1111/mice.12494]
Kuok S C and Yuen K V. 2020b. Broad learning system for nonparametric modeling of clay parameters. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 6(2): 04020024 [DOI: 10.1061/ajrua6.0001066http://dx.doi.org/10.1061/ajrua6.0001066]
Liu C H and Niu R Y. 2013. Object-based precipitation verification method and its application. Meteorological Monthly, 39(6): 681-690
刘凑华, 牛若芸. 2013. 基于目标的降水检验方法及应用. 气象, 39(6): 681-90 [DOI: 10.7519/j.issn.1000-0526.2013.06.003http://dx.doi.org/10.7519/j.issn.1000-0526.2013.06.003]
Liu S X. 2021. Precipitation Detection Algorithm over Land for FY3C MWHS-2 and its Application in LAPS-WRF Assimilation System. Nanjing: Nanjing University of Information Science and Technology
刘淑贤. 2021. FY3C MWHS-2的陆地降水检测算法研究及其在LAPS-WRF同化系统中的应用. 南京: 南京信息工程大学 [DOI: 10.27248/d.cnki.gnjqc.2021.000012http://dx.doi.org/10.27248/d.cnki.gnjqc.2021.000012]
Liu X N, Zheng Q A, Liu R, Sletten M A and Duncan J H. 2017. A model of radar backscatter of rain-generated stalks on the ocean surface. IEEE Transactions on Geoscience and Remote Sensing, 55(2): 767-776 [DOI: 10.1109/tgrs.2016.2614897http://dx.doi.org/10.1109/tgrs.2016.2614897]
Liu X N, Zheng Q A, Liu R, Wang D, Duncan J H and Huang S J. 2016. A study of radar backscattering from water surface in response to rainfall. Journal of Geophysical Research: Oceans, 121(3): 1546-1562 [DOI: 10.1002/2015jc010975http://dx.doi.org/10.1002/2015jc010975]
Lü B Q. 2020. Research on Rainfall Detection Technology Inversion Method based on Marine Radar Image. Harbin: Harbin Engineering University
吕博群. 2020. 基于航海雷达图像反演的降雨检测技术研究. 哈尔滨: 哈尔滨工程大学 [DOI: 10.27060/d.cnki.ghbcu.2020.001497http://dx.doi.org/10.27060/d.cnki.ghbcu.2020.001497]
Ma C Y. 2019. Research on Dynamic Detection Algorithm of Precipitation Clouds. Dalian: Dalian Maritime University
马春野. 2019. 降水云团动态检测算法研究. 大连: 大连海事大学 [DOI: 10.26989/d.cnki.gdlhu.2019.000751http://dx.doi.org/10.26989/d.cnki.gdlhu.2019.000751]
Melsheimer C, Alpers W and Gade M. 2001. Simultaneous observations of rain cells over the ocean by the synthetic aperture radar aboard the ERS satellites and by surface-based weather radars. Journal of Geophysical Research: Oceans, 106(C3): 4665-4677 [DOI: 10.1029/2000jc000263http://dx.doi.org/10.1029/2000jc000263]
Tao R Z. 2021. Research on Cloud Detection and Precipitation Extrapolating Forecast in Tibet based on FY-4A Satellite Images Nanjing: Nanjing University of Information Science and Technology
陶润喆. 2021. 基于风云4号卫星图像的西藏地区云检测和降水外推预报研究. 南京: 南京信息工程大学 [DOI: 10.27248/d.cnki.gnjqc.2021.000052http://dx.doi.org/10.27248/d.cnki.gnjqc.2021.000052]
Wang C, Mouche A, Tandeo P, Stopa J E, Longépé N, Erhard G, Foster R C, Vandemark D and Chapron B. 2019. A labelled ocean SAR imagery dataset of ten geophysical phenomena from Sentinel-1 wave mode. Geoscience Data Journal, 6(2): 105-115 [DOI: 10.1002/gdj3.73http://dx.doi.org/10.1002/gdj3.73]
Xu F, Li X F, Wang P, Yang J S, Pichel W G and Jin Y Q. 2015. A backscattering model of rainfall over rough sea surface for Synthetic Aperture Radar. IEEE Transactions on Geoscience and Remote Sensing, 53(6): 3042-3054 [DOI: 10.1109/tgrs.2014.2367654http://dx.doi.org/10.1109/tgrs.2014.2367654]
Zhang Y and Yuen K V. 2021. Crack detection using fusion features-based broad learning system and image processing. Computer-Aided Civil and Infrastructure Engineering, 36(12): 1568-1584 [DOI: 10.1111/mice.12753http://dx.doi.org/10.1111/mice.12753]
Zhou X, Yang X F, Li Z W, Yu Y and Ma S. 2012. Rain effect on C-band scatterometer wind measurement and its correction. Acta Physica Sinica, 61(14): 149202
周旋, 杨晓峰, 李紫薇, 于旸, 马胜. 2012. 降雨对C波段散射计测风的影响及其校正. 物理学报, 61(14): 149202 [DOI: 10.7498/aps.61.149202http://dx.doi.org/10.7498/aps.61.149202]
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
湖南大学 电气与信息工程学院
加拿大海洋渔业部 贝德福德海洋研究所, 达特茅斯 NSBY24A2
连云港市气象局
浙江省气候中心
自然资源部第二海洋研究所 卫星海洋环境动力学国家重点实验室
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