Man–computer interactive method on cloud classification based on bispectral satellite imagery
暂无分享,去创建一个
A bispectral cloud classification method based on man-computer interactive way, i.e. a unit feature space classification method (UFSCM), has been presented in this paper. Apart from land and water, six types of clouds including cumulonimbus, multilayer cloud system, thin / think cirrus, middle and low level clouds are recognized. The method has been tested by using more than two hundred samples, with total accuracy reaching 87.1 %.
[1] Li Jun,et al. Computer identification of multispectral satellite cloud imagery , 1990 .
[2] C. Platt,et al. On the Bispectral Method for Cloud Parameter Determination from Satellite VISSR Data: Separating Broken Cloud and Semitransparent Cloud. , 1983 .
[3] Liu Changsheng,et al. Extraction of Cloud Parameters from Bispectral Measurements , 1987 .
[4] Zhou Fengxian,et al. Automatic classification and compression of GMS cloud imagery in heavy rainfall monitoring application , 1992 .