Volume XLII-3/W9
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W9, 219–226, 2019
https://doi.org/10.5194/isprs-archives-XLII-3-W9-219-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLII-3/W9, 219–226, 2019
https://doi.org/10.5194/isprs-archives-XLII-3-W9-219-2019
© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

  25 Oct 2019

25 Oct 2019

DECISION TREE CLOUD DETECTION ALGORITHM BASED ON FY-4A SATELLITE DATA

Z. F. Yu, W. H. Ai, Z. H. Tan, and W. Yan Z. F. Yu et al.
  • College of Meteorology and Oceanography, National University of Defense Technology, Nanjing, China

Keywords: FY-4A, Cloud detection, Decision tree, on-board processing, Infrared, Visible light

Abstract. In order to study the on-board processing technology of meteorological satellites, a decision tree cloud detection algorithm is proposed by taking FY-4A satellite data as an example. According to the channel setting of the Advanced Geosynchronous Radiation Imager (AGRI) on FY-4A satellite, the 0.65 μm, 1.375 μm, 3.75 μm, and 10.7 μm bands are selected as the cloud detection channels, and the reflectance, brightness temperature or bright temperature difference of the four channels are used as the cloud detection indicators, the thresholds of the four cloud detection indicators are obtained through statistics. On this basis, the decision tree cloud detection model is constructed and validated using FY-4A satellite data. The results show that the algorithm is simple, convenient and efficient, and the overall effect of cloud detection is good. It is an effective way for meteorological satellite cloud detection on-board processing technology.