摘要
针对地球观测系统/中分辨率成像光谱仪影像资料中的云检测工作,提出了基于支持向量机SVMs(support vector machines)的遥感影像分类方法。分析了云检测过程中的特征提取和选择,建立了基于支持向量机的遥感影像分类模型,并针对陆地、海洋2种不同的下垫面进行了云检测试验。云检测结果中,云与陆地、水体、积雪准确地区分开来。结果表明,特征选择对云检测起到了积极的作用,同时也证明了支持向量机方法在遥感影像分类中的优势。
A new method of remote sensing image classification was introduced based on the support vector machines by using the images from earth observing system/moderate resolution imaging spectroradiometer. The process of feature selection during cloud detection was analyzed. The new method was established by building a model of remote sensing image classification based on SVMs that was used in different surfaces of land and sea images. Clouds were discriminated from land, water and snow in the preliminary results. The established scheme based on SVMs method proves to be effective in remote sensing image classification.
出处
《解放军理工大学学报（自然科学版）》
EI
北大核心
2009年第2期191-194,共4页
Journal of PLA University of Science and Technology(Natural Science Edition)
基金
国家重点基础研究规划资助项目(2006CB400505)
国家自然科学基金资助项目(40675040)
关键词
支持向量机
中分辨率成像光谱仪
云检测
support vector machines
MODIS（moderate resolution imaging spectroradiometer） cloud detection
