The Cloud Detection Study of MODIS Based on HSV Color Clustering Algorithm

Abstract-The cloud is a primary factor for influencing the optical satellite remote sensing image quality. Cloud detection is the necessary link in the image data analysis processing. MODIS data is the optimal remote sensing data source for detecting cloud .The paper puts forward a new cloud detection algorithm of MODIS data with K-mean clustering and HSV color model. The proposed method just makes use of three bands of the MODIS data. In addition it does not need any priori knowledge .It is suitable for the MODIS daytime Data of the different season and different earth´s surface, especially snow, dessert, plateau and ocean. It is mature, simple, fast and automatic without manual intervention. The experimental result by the HSV color clustering cloud detection is very ideal.