面向对象的遥感影像分类在地理国情普查中的应用

遥感技术因其较高的精确性和时效性已成为监测土地利用变化的重要手段。地理国情普查是一项非常重要的调查工程,起着为各级领导干部科学决策提供数据支持的作用。遥感影像解译是整个普查工作中的重要环节,而且工作量巨大。目前,地理国情普查的遥感影像解译环节主要采用人工解译和计算机自动解译相结合的方法,以提高效率。本文主要介绍一种面向对象分类的遥感影像自动解译方法,通过东莞市TM影像实验,总结该方法的解译特点,并介绍其在地理国情普查中的应用。

Application of Object-oriented Remote-sensing Image Classification in the Geographical National Census

Remote sensing technology is an rapid and accurate method which is becoming more and more important in monitoring land use change.Geography Census survey is a very important project,plays a important role in providing data support for government in decision-making.Remote sensing image interpretation is an important part of Geographical National Census.Currently,remote sensing image interpretation in the census usually combines the artificial interpretation and computer automatic interpretation,in order to improve efficiency.This paper describes an object-oriented classification method of remote sensing image.In the case of Dongguan TM image experiment,we summarizes the characteristics of this interpretation method,and describes the application of this method in the Geographical National Census.