摘要
提出了一种在对遥感影像分类的基础上进行地图更新的方法,讨论了利用高分辨率遥感影像,通过不 同空间分辨率和光谱分辨率的影像进行融合,利用合适的高通滤波对影像进行边缘检测,构建一个三层的 MLP分类器对影像进行分类,提取城市建筑物与道路信息,并在此分类基础上通过对现有地图的叠加来实现 地图的更新。实验结果表明,基于影像融合,利用较少数量的训练样本也能生成具有较高精度的分类图,利用 分类结果图进行地图更新能取得令人满意的效果。
On the basis of remote sensing image classification, this paper introduces a method for updating urban maps through high resolution remote sensing image. The RGB-HIS transformation technique is considered to merge spectral and spatial information, an appropriate filter for the edge extraction has been used. An MLP classifier is trained to produce a labeled image, focusing on urban buildings and roads, with great accuracy in test even if a limited training set is used. The result reveals a good feasibility of the classified image for monitoring the presence of changes in urban areas.
出处
《武汉大学学报(信息科学版)》
EI
CSCD
北大核心
2005年第2期105-109,共5页
Geomatics and Information Science of Wuhan University
基金
国家自然科学基金资助项目(60175022)
关键词
影像融合
影像分类
地图更新
image fusion
image classification
map revision