摘要
本文提出MSS数据波谱形态相关模型和直方图决策树分类法。这一方法从数据直方图和矫正的波谱形态曲线中提取模型参数、直方图特征参数、各波段相关系数参数,同时提取波谱反转特性参数,然后利用直方图决策树分类器进行计算机判决分类。将该方法应用于咸宁市县级土地利用状况调查中,取得了满意效果。
An MSS spectral form correlation model and a histogram decision tree classifier are presented. A set of parameters, the model parameters, histogram feature parameters, all bands correlation coefficients and spectral inversion features, are derived from MSS data histograms and the calibrated spectral curves. A histogram decision tree computer classifier is used to classify these parameters. The model and relevant method have been successfully used in the survey of land exploitation in Xianling District of Hubei Province.
出处
《华中理工大学学报》
CSCD
北大核心
1990年第5期139-146,共8页
Journal of Huazhong University of Science and Technology
关键词
土地利用调查
遥感数据
计算机分类
Land survey
Remote sensing data
Computer classification
Statistical method
