A New Application of Unsupervised Learning to Nighttime Sea Fog Detection

다국어 초록 (Multilingual Abstract)

This paper presents a nighttime sea fog detection algorithm incorporating unsupervised learning technique. The algorithm is based on data sets that combine brightness temperatures from the 3.7 μm and 10.8 μm channels of the meteorological imager (MI) onboard the Communication, Ocean and Meteorological Satellite (COMS), with sea surface temperature from the Operational Sea Surface Temperature and Sea Ice Analysis (OSTIA). Previous algorithms generally employed threshold values including the brightness temperature difference between the near infrared and infrared. The threshold values were previously determined from climatological analysis or model simulation. Although this method using predetermined thresholds is very simple and effective in detecting low cloud, it has difficulty in distinguishing fog from stratus because they share similar characteristics of particle size and altitude. In order to improve this, the unsupervised learning approach, which allows a more effective interpretation from the insufficient information, has been utilized. The unsupervised learning method employed in this paper is the expectation–maximization (EM) algorithm that is widely used in incomplete data problems. It identifies distinguishing features of the data by organizing and optimizing the data. This allows for the application of optimal threshold values for fog detection by considering the characteristics of a specific domain. The algorithm has been evaluated using the Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) vertical profile products, which showed promising results within a local domain with probability of detection (POD) of 0.753 and critical success index (CSI) of 0.477, respectively.

참고문헌 (Reference)

1 Park, H.-S., "offog and the low stratus cloud at night using derived dual channel difference of NOAA/AVHRR data" 1997

2 Kim, S.-W., "Validation of aerosol and cloud layer structures from the spaceborne lidar CALIOP using a ground-based lidar in Seoul, Korea" 8 : 3705-3720, 2008

3 Papin, C., "Unsupervisedsegmentationoflow clouds from infrared METEOSAT images based on a contextual spatio–temporal labeling approach" 40 : 104-114, 2002

4 Turk, F., "Toward improving estimates of remotely sensed precipitation with MODIS/AMSR-E blended data techniques" 43 : 1059-1069, 2005

5 Gultepe, I., "The fog remote sensing and modelingfield project" 90 : 341-359, 2009

6 Lee, T.F., "Stratus and fog products using GOES-8-9 3.9-μm data" 12 : 664-677, 1997

7 Cho, Y.-K, "Sea fog around the Korean peninsula" 39 : 2473-2479, 2000

8 Hunt, G.E., "Radiative properties of terrestrial clouds at visible and infrared thermal window wavelengths" 99 : 346-369, 1973

9 Pankiewicz, G.S., "Pattern recognition techniques for the identification of cloud and cloud systems" 2 : 257-271, 1995

10 Bendix, J., "Operational detection of fog in the alpine region by means of advanced very high resolution radiometer (AVHRR) imagery of NOAA satellites" 307-312, 1991

11 Stark, J.D., "OSTIA: an operational, high resolution, real time, global sea surface temperature analysis system" IEEE 1-4, 2007

12 Schreiner, A.J., "Notes and correspondence; A multispectral technique for detecting lowlevel cloudiness near sunrise" 24 : 1800-1810, 2007

13 Bendix, J., "New perspectives in remote sensing of fog and low stratus-TERRA/AQUA-MODIS and MSG" 2004

14 Dempster, A.P., "Maximum likelihood from incomplete data via the EM algorithm" 39 : 1-38, 1977

15 박형민, "MTSAT 적외채널과 AMSR 마이크로웨이브채널의 결합을 이용한 한반도 주변의 해무 탐지" 한국기상학회 22 (22): 163-174, 2012

16 d’Entremont, R.P., "Low-and midlevel cloud analysis using nighttime multispectralimagery" 25 : 1853-1869, 1986

17 d’Entremont, R.P., "Interpreting meteorological satellite imagesusingacolor-compositetechnique" 68 : 762-776, 1987

18 Ellrod, G. P., "Inferring low Cloud Base heights at night for aviation using satellite infrared and surface temperature data" 164 : 1193-1205, 2007

19 Cha, Y.-M, "Impacts of the high-Resolution Sea surface temperature distribution on modeled snowfall formation over the Yellow Sea during a cold-air outbreak" 26 : 487-503, 2011

20 Calvert, C., "GOES-R advanced baseline imager (ABI) algorithm theoretical basis document for low cloud and fog version 1.0"

21 Whiffen, B., "Fog: impact on aviation and goals for meteorological prediction" Environment Canada and WMO, St. John’s 525-528, 2001

22 Gultepe, I., "Fog research: a review of past achievements and future perspectives" 164 : 1121-1159, 2007

23 Bendix, J., "Fog detection with TERRA-MODIS and MSG-SEVIRI" 427-435, 2003

24 Li, J., "Fog detection over China’s Adjacent Sea area by using the MTSAT geostationary satellite data" 5 (5): 128-133, 2012

25 이정림, "Fog Detection Using Geostationary Satellite Data: Temporally Continuous Algorithm" 한국기상학회 47 (47): 113-122, 2011

26 McLachlan, G., "Finite Mixture Models" Wiley 2000

27 Zhang, Z., "EM algorithms for Gaussian mixtureswithsplit-and-mergeoperation" 36 : 1973-1983, 2003

28 Cermak, J., "Dynamical nighttime fog/low stratus detection based on Meteosat SEVIRI data: a feasibility study" 164 : 1179-1192, 2007

29 Gentemann, C. L., "Diurnal signals in satellite sea surface temperature measurements" 30 : 1140-, 2003

30 Kawai, Y., "Diurnal Sea surface temperature variation and its impact on the atmosphere and ocean: a review" 63 : 721-744, 2007

31 National Institute of Meteorological Research, "Development of Meteorological Data Processing System for Communication"

32 Gao, S., "Detection of nighttime sea fog/stratus over the Huang-Hai Sea using MTSAT-1R IR data" 28 : 23-35, 2009

33 Eyre, J.R., "Detection of fog at night usingadvanced very highresolution radiometer (AVHRR)imagery" 113 : 266-271, 1984

34 Cermak ,J., "Detectinggroundfogfromspace – amicrophysicsbased approach" 32 : 3345-3371, 2011

35 Xie, J., "Assessment and inter-comparison of five highresolution sea surface temperature products in the shelf and coastal seas around China" 28 : 1286-1293, 2008

36 Saunders, R. W., "An improved method for detecting clear sky and cloudy radiances from AVHRRdata" 9 : 123-150, 1988

37 Hartigan, J.A., "Algorithm AS 136: a k-means clustering algorithm" 28 : 100-108, 1979

38 Ellrod, G.P., "Advances in the detection and analysis offog at night using GOES multispectral infrared imagery" 10 : 606-619, 1995

39 Cermak, J., "A novel approach to fog/low stratus detection using Meteosat 8 data" 87 : 279-292, 2008

40 Cermak, J., "A new approach to fog detection using SEVIRI and MODIS data" 2004

41 Ahn, M.-H., "A new algorithm for sea fog/ stratus detection using GMS-5 IR data" 20 : 899-913, 2003

42 Wu, D., "A method of detecting sea fogs using CALIOP data and its application to improve MODIS-based sea fog detection" 153 : 88-94, 2015

43 Zhang, S., "A comprehensive dynamic threshold algorithm for Daytime Sea fog retrieval over the Chinese adjacent seas" 170 : 1931-1944, 2013

44 허기영, "A Remote Sensed Data Combined Method for Sea Fog Detection" 대한원격탐사학회 24 (24): 1-16, 2008