Research on Land Use Classification in a Certain Area in Hunan Province Based on High-Resolution Satellite Images
With the rapid development of remote sensing technology,the emergency of more and more high-resolution remote sensing images creates favorable conditions for quickly and accurately obtaining ground informa-tion.The object-oriented classification method takes into account more classification features when extracting ground-object information,and can be combined with geoscience knowledge and other thematic features,so that the classification process is closer to the cognitive process of human beings,which has become one of the main-stream directions of research on land use information extraction.This study is based on high-resolution No.2 im-ages,explores the selection method of optimal parameters for multi-scale segmentation,constructs image classifica-tion feature space and optimizes it.It extracts land use information based on a multi-level classification system and conducts accuracy evaluation.It subdivides construction land information through spatial big data and analyzes its spatial distribution characteristics.
NETL
NSTL
万方数据
