结合FY-4A卫星及随机森林的日间沿海海雾识别模型的研究-海洋预报

结合FY-4A卫星及随机森林的日间沿海海雾识别模型的研究

单位：1. 江苏省气象信息中心, 江苏南京 210041;
2. 江苏省气象服务中心, 江苏南京 210041;
3. 江苏省气象台, 江苏南京 210041

分类号：P732

出版年·卷·期（页码）：2022·39·第三期（83-93）

摘要：

利用2019年8月—2021年7月期间的FY-4A卫星数据,结合江苏及周边地区自动气象站同期的能见度观测数据,建立了包括有海雾时次及非海雾的卫星像素集,利用随机森林算法构建了海雾识别模型,实现对江苏及周边区域的日间海雾识别。检验结果表明:与基于阈值法的海雾识别模型相比,训练得到的随机森林海雾识别模型具有较高的识别精度,该模型平均命中率、平均临界成功指数和平均误报率分别为83.46%、79.46%和5.7%,均优于阈值法的结果。两种识别模型对2021年4月12日发生在黄渤海区域海雾天气个例识别结果的对比表明,随机森林海雾识别模型能够更好地识别出发生海雾的区域。

Using the FY-4A satellite data from August 2019 to July 2021 combined with the visibility observation data of automatic weather stations in Jiangsu province and surrounding areas in the same period, a satellite pixel set including sea fog and non-sea fog is established, and the daytime sea fog identification in Jiangsu and surrounding areas is realized based on a sea fog recognition model that is constructed using random forest algorithm. The validation results show that the random forest sea fog recognition model trained in this paper has higher recognition accuracy compared with the sea fog recognition model based on threshold method, and the average hit rate, average critical success index and average false positive rate of the model are 83.46%, 79.46%and 5.7%, respectively. At the same time, by comparing the identification results of the two identification models for a sea fog weather case in the Yellow Sea and Bohai Sea on April 12, 2021, it shows that the random forest sea fog identification model can better identify the sea fog area.

参考文献：

[1] 王宏斌,张志薇,刘端阳,等.基于葵花8号新一代静止气象卫星的夜间雾识别[J].高原气象,2018,37(6):1749-1764.Wang H B,Zhang Z W,Liu D Y,et al.Detection of fog at night by using the new geostationary satellite Himawari-8[J].Plateau Meteorology,2018,37(6):1749-1764.
[2] 郑新江.黄海海雾的卫星云图特征分析[J].气象,1988,14(6):7-9,65.Zheng X J.On satellite imagery features of sea fogs over the Yellow Sea[J].Meteorological Monthly,1988,14(6):7-9,65.
[3] Ellrod G P.Advances in the detection and analysis of fog at night using GOES multispectral infrared imagery[J].Weather and Forecasting,1995,10(3):606-619.
[4] 鲍献文,王鑫,孙立潭,等.卫星遥感全天候监测海雾技术与应用[J].高技术通讯,2005,15(1):101-106.Bao X W,Wang X,Sun L T,et al.The weatherproof detection system of sea fog by remote sensing and its applications[J].Chinese High Technology Letters,2005,15(1):101-106.
[5] 何月,张小伟,杜惠良,等.利用静止气象卫星监测浙江海上大雾[J].遥感技术与应用,2015,30(3):599-606.He Y,Zhang X W,Du H L,et al.Monitoring sea fog of Zhejiang from geostationary meteorological satellite data[J].Remote Sensing Technology and Application,2015,30(3):599-606.
[6] Shang H Z,Chen L F,Letu H,et al.Development of a daytime cloud and haze detection algorithm for Himawari-8 satellite measurements over central and eastern China[J].Journal of Geophysical Research:Atmospheres,2017,122(6):3528-3543.
[7] 张培,吴东.基于Himawari-8数据的日间海雾检测方法[J].大气与环境光学学报,2019,14(3):211-220.Zhang P,Wu D.Daytime sea fog detection method using Himawari-8 data[J].Journal of Atmospheric and Environmental Optics,2019,14(3):211-220.
[8] 衣立.黄海海雾/层云的空间分布及云底高度遥感方法研究[D].青岛:中国海洋大学,2015.Yi L.Spatio-temporal detection of sea fog/stratus and cloud base height over yellow sea with satellite data-A feasibility study[D].Qingdao:Ocean University of China,2015.
[9] Wang B,Dong L L,Zhao M,et al.An infrared maritime target detection algorithm applicable to heavy sea fog[J].Infrared Physics & Technology,2015,71:56-62.
[10] 张春桂,林炳青.基于FY-2E卫星数据的福建沿海海雾遥感监测[J].国土资源遥感,2018,30(1):7-13.Zhang C G,Lin B Q.Application of FY-2E data to remote sensing monitoring of sea fog in Fujian coastal region[J].Remote Sensing for Land & Resources,2018,30(1):7-13.
[11] 孙艺,杨悦,甄晴.CALIPSO卫星资料的春夏季黄海海雾高度特征分析[J].海洋预报,2020,37(3):54-61.Sun Y,Yang Y,Zhen Q.The characteristics of the top height of sea fog over the Yellow Sea in spring and summer based on CALIPSO satellite data[J].Marine Forecasts,2020,37(3):54-61.
[12] 于海鹏,田付友.GOES9云图在黄海海雾区域识别中的应用[J].海洋预报,2010,27(3):23-29.Yu H P,Tian F Y.Application of GOES9 cloud images in sea tog in the Yellew Sea[J].Marine Forecasts,2010,27(3):23-29.
[13] Kim D,Park M S,Park Y J,et al.Geostationary ocean color imager (GOCI) marine fog detection in combination with Himawari-8 based on the decision tree[J].Remote Sensing,2020,12(1):149,doi:10.3390/rs12010149.
[14] Shin D,Kim J H.A new application of unsupervised learning to nighttime sea fog detection[J].Asia-Pacific Journal of Atmospheric Sciences,2018,54(4):527-544.
[15] 许赟,许艾文.基于随机森林的遥感影像云雪雾分类检测[J].国土资源遥感,2021,33(1):96-101.Xu Y,Xu A W.Classification and detection of cloud,snow and fog in remote sensing images based on random forest[J].Remote Sensing for Land & Resources,2021,33(1):96-101.
[16] 姜红,何清,曾晓青,等.基于随机森林和卷积神经网络的FY-4A号卫星沙尘监测研究[J].高原气象,2021,40(3):680-689.Jiang H,He Q,Zeng X Q,et al.Sand and dust monitoring using FY-4A satellite data based on the random forests and convolutional neural networks[J].Plateau Meteorology,2021,40(3):680-689.
[17] 张环宇,唐伯惠.融合物理机理与随机森林算法的FY-4A AGRI数据晴空大气可降水量遥感反演[J].遥感学报,2021,25(8):1836-1847.Zhang H Y,Tang B H.Remote sensing retrieval of total precipitable water under clear-sky atmosphere from FY-4A AGRIdata by combining physical mechanism and random forest algorithm[J].National Remote Sensing Bulletin,2021,25(8):1836-1847.
[18] 柳青青,孟朔羽,徐茗,等.随机森林反演卫星遥感海表面盐度研究[J].武汉大学学报·信息科学版,2021,doi:10.13203/j.whugis20210153.Liu Q Q,Meng S Y,Xu M,et al.Satellite sea surface salinity retrieval using random forest model[J].Geomatics and Information Science of Wuhan University,2021,doi:10.13203/j.whugis20210153.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】