To study the classification of satellite images availability, the description about the structural characteristics of clouds by coverage rate, thickness and fragment of clouds based on image information was proposed. The influence of these structural characteristics to satellite images availability was studied,and the orthogonal space of availability assessment was built. The characteristics of approximate linear and monotone between cloud structures and availability of satellite images were analyzed. The 500 satellite images categorized by people trained professionally were used to test the classifier. The result showed that 95% of the images classified by these three characteristics were the same as those classified by people trained professionally. The theoretical and technical support to build availability of satellite image evaluation model and automatically classifies satellite images according to the needs of users was provided by this description.
Cao Qiong.Research on availability evaluation intelligent method of satellite remote sensing image .Beijing:School of Automation Science and Electrical Engineering,Beijing University of Aeronautics and Astronautics,2008(in Chinese)
Tong Yubing,Zhang Qishan,Chang Qing,et al.Image quality assessing model by using neural network and support vector machine[J].Journal of Beijing University of Aeronautics and Astronautics,2006,32(9):1031-1034(in Chinese)
Yang Wei,Zhao Yan,Xu Dong.Method of image quality assessment based on human visual system and structural similarity[J].Journal of Beijing University of Aeronautics and Astronautics,2008,34(1):1-4(in Chinese)
Ellen M,Hothem D,John M I.The multispectral imagery interpretability rating scale (MS IIRS) //Lewis,Anthony J[J].ASPRS/ACSM Annual Convention & Exposition Technical Papers.Bethesda:ASPRS/ACSM.1996,1:300-310