An FY-4A/AGRI cloud detection model based on the naive Bayes algorithm
-
摘要: 针对风云四号A星(FY-4A)中多通道扫描成像辐射计(advanced geosynchronous radiation imager,AGRI)云检测问题,提出了一种基于朴素贝叶斯算法的全自动云检测方法。使用朴素贝叶斯算法作为核心结构,基于光学载荷基本云检测原理选择合适的红外通道作为特性分类器参数,可保证日夜云检测一致性,同时针对不同的地表类型和不同月份分别分类训练构建,最终得到基于朴素贝叶斯算法的云检测模型。针对FY-4A/AGRI数据生成了7种经典的云检测特征和1种基于红外合成图像特征的贝叶斯分类器,经过2019年国家卫星气象中心业务云检测产品的学习测试验证,在陆地、沙漠、浅水和深海的召回率(probability of detection,POD)达到98%以上,积雪POD达到80%,南北极POD达到80%以上。将检测结果与国家卫星气象中心业务系统云检测结果进行比较,全年月度平均POD均高于98%,误判率(false alarm ratio,FAR)低于5%,Kuipers评分(Kuiper's skill score,KSS)均高于90%。
-
关键词:
- FY-4A/AGRI /
- 朴素贝叶斯 /
- 云检测 /
- POD
Abstract: This study developed an automatic cloud detection method based on the naive Bayes algorithm for the cloud detection of the advanced geosynchronous radiation imager (AGRI) aboard the FY-4A satellite. In this method, the naive Bayes algorithm serves as the core structure, and appropriate infrared channels are selected as the parameters of the characteristic classifier according to the basic cloud detection principle of optical payload to ensure the consistency of cloud detection between day and night. After the classified training and construction for different surface types and different months, a cloud detection model based on the naive Bayes algorithm was finally established. Moreover, the classifier for FY-4A/AGRI data used in the method was established considering seven typical cloud detection characteristics and one characteristic based on the infrared composite images. As indicated by the learning tests and verification using the business cloud detection product of the National Satellite Meteorological Center (NSMC) in 2019, the classifier yielded a probability of detection (POD) greater than 98% for land, desert, shallow water, and deep sea, greater than 80% for snow cover, and greater than 80% for North and South poles. The comparison between the cloud detection results of this study and those obtained using the NSMC business system showed that the cloud detection results of this study had an average monthly POD of the whole year greater than 98%, a false alarm ratio (FAR) less than 5%, and all Kuiper’s skill scores (KSSs) greater than 90%.-
Key words:
- FY-4A/AGRI /
- Naive Bayes Algorithm /
- cloud detection /
- POD
-
-
[1] 林晔. 大气探测学教程[M]. 北京: 气象出版社, 1995.
[2] Lin Y. Course of atmospheric exploration[M]. Beijing: China Meteorological Press, 1995.
[3] 杨亦萍, 董晓刚, 戴聪明, 等. 利用MODIS数据对北极夏季卷云特性的研究[J]. 红外与激光工程, 2016, 45(4):30-37.
[4] Yang Y P, Dong X G, Dai C M, et al. Cirrus clouds properties in the arctic in summer based on MODIS data[J]. Infrared and Laser Engineering, 2016, 45(4):30-37.
[5] 秦永, 付仲良, 周凡, 等. 一种利用TM影像自动检测云层的方法[J]. 武汉大学学报(信息科学版), 2014, 39(2):234-238.
[6] Qin Y, Fu Z L, Zhou F, et al. A method of automatic cloud detection using TM image[J]. Geomatics and Information Science of Wuhan University, 2014, 39(2):234-238.
[7] 赵忠明, 朱重光. 遥感图象中薄云的去除方法[J]. 环境遥感, 1996, 11(3):195-199.
[8] Zhao Z M, Zhu C G. Approach to removing cloud cover from satellite imagery[J]. Remote Sensing of Environment, 1996, 11(3):195-199.
[9] Rossow W B, Schiffer R A. ISCCP cloud data products[J]. Bulletin of the American Meteorological Society, 1991, 72(1):2-20.
[10] Seze G, Rossow W B. Time-cumulated visible and infrared radiance histograms used as descriptors of surface and cloud variations[J]. International Journal of Remote Sensing, 1991, 12(5):877-920.
[11] Rossow W B, Garder L C. Cloud detection using satellite measurements of infrared and visible radiances for ISCCP[J]. Journal of Climate, 1993, 6(12):2341-2369.
[12] 刘健. FY-2云检测中动态阈值提取技术改进方法研究[J]. 红外与毫米波学报, 2010, 29(4):288-292.
[13] Liu J. Improvement of dynamic threshold value extraction technic in FY-2 cloud detection[J]. Journal of Infrared and Millimeter Waves, 2010, 29(4):288-292.
[14] Saunders R W, Kriebel K T. An improved method for detecting clear sky and cloudy radiances from AVHRR data[J]. International Journal of Remote Sensing, 1988, 9(1):123-150.
[15] Kriebel K T, Saunders R W, Gesell G. Optical properties of clouds derived from fully cloudy AVHRR pixels[J]. Beitrage zur Physik der freien Atmosphare, 1989, 62:165-171.
[16] Gesell G. An algorithm for snow and ice detection using AVHRR data an extension to the APOLLO software package[J]. International Journal of Remote Sensing, 1989, 10:4-5,897-905.
[17] Stowe L L, McClain E P, Carey R, et al. Global distribution of cloud cover derived from NOAA/AVHRR operational satellite data[J]. Advances in Space Research, 1991, 11(3):51-54.
[18] Stowe L L, Vemury S K, Rao A V. AVHRR clear sky radiation data sets at NOAA/NESDIS[J]. Advance in Space Research, 1994, 14(1):113-116.
[19] Menzel W P, Wylie D P, Strabala K I. Seasonal and diurnal changes in cirrus clouds as seen in four years of observations with the VAS[J]. Journal of Applied Meteorology, 1992, 31(4):370-385.
[20] Wylie D P, Menzel W P, Woolf H M, et al. Four years of global cirrus cloud statistics using HIRS[J]. Journal of Climate, 1994, 7(12):1972-1986.
[21] Shan N, Zheng T Y, Wang Z S. High-speed and high-accuracy algorithm for cloud detection and its application[J]. Journal of Remote Sensing, 2009, 13(6):1138-1146.
[22] Goodman A H, Henderson-Sellers A. Cloud detection and analysis:A review of recent progress[J]. Atmospheric Research, 1988, 21(3-4):203-228.
[23] 韩成鸣, 李耀东, 史小康. 云分析预报方法研究进展[J]. 地球科学进展, 2015, 30(4):505-516.
[24] Han C M, Li Y D, Shi X K. Overview of researches on cloud analysis and prediction methods[J]. Advances in Earth Science, 2015, 30(4):505-516.
[25] 孙汝星, 范荣双. 基于支持向量机的多特征融合影像云检测[J]. 测绘与空间地理信息, 2018, 41(6):153-156,160.
[26] Sun R X, Fan R S. Multi-feature fusion image cloud detection based on SVM[J]. Geomatics and Spatial Information Technology, 2018, 41(6):153-156,160.
[27] 关峥嵘, 谭毅华, 田金文. 一种快速的遥感卫星图像云检测算法. 计算机与数字工程, 2019, 47(11):2808-2811.
[28] Guan Z R, Tan Y H, Tian J W. A fast cloud detection method in remote sensing satellite images[J]. Computer and Digital Engineering, 2019, 47(11):2808-2811.
[29] 瞿建华, 鄢俊洁, 薛娟, 等. 基于深度学习的FY3D/MERSI和EOS/MODIS云检测模型研究[J]. 气象与环境学报, 2019, 35(3):87-93.
[30] Qu J H, Yan J J, Xue J, et al. Research on the cloud detection model of FY3D /MERSI and EOS /MODIS based on deep learning[J]. Journal of Meteorology and Environment, 2019, 35(3):87-93.
[31] 董瑶海. 风云四号气象卫星及其应用展望[J]. 上海航天, 2016, 33(2):1-8.
[32] Dong Y H. Fy-4 meteorological satellite and its application prospect[J]. Aerospace Shanghai, 2016, 33(2):1-8.
[33] 陆风, 张晓虎, 陈博洋, 等. 风云四号气象卫星成像特性及其应用前景[J]. 海洋气象学报, 2017, 37(2):1-12.
[34] Lu F, Zhang X H, Chen B Y, et al. FY-4 geostationary meteorological satellite imaging characteristics and its application prospects[J]. Journal of Marine Meteorology, 2017, 37(2):1-12.
[35] 杨军. 气象卫星及其应用[M]. 北京: 气象出版社. 2012.
[36] Yang J. Meteorological satellites and their applications[M]. Beijing: China Meteorological Press, 2012.
[37] Ackerman S A, Smith W L, Revercomb H E, et al. The 27-28 October 1986 FIRE IFO cirrus case study:Spectral properties of cirrus clouds in the 8-12μm[J]. Monthly Weather Review, 1990, 118(11),2377-2388.
[38] Miller S D, Lindsey D T, Seaman C J, et al. GeoColor:A blending technique for satellite imagery[J]. Journal of Atmospheric and Oceanic Technology, 2020, 37(3):429-448.
-
计量
- 文章访问数: 430
- PDF下载数: 50
- 施引文献: 0
下载: