结合GIS数据和遥感影像的地类图斑更新研究

为提高地类图斑的更新准确率,提出一种结合GIS数据和高分辨率遥感影像的土地利用现状库中局部地类图斑快速更新的方法。该方法在遥感影像和地类图斑数据坐标配准套和的前提下,首先采用面向对象的影像分析方法对遥感影像进行分类,并利用已有的地类图斑数据作为分类的先验知识,以提高影像分类的精度;然后将影像分类后的矢量结果与地类图斑数据进行布尔判断确定变化区,对变化区进行变化类型更新。实验结果表明提出的方法能准确的更新地类图斑,是一种可行的土地利用现状数据更新方法。

Research of Class Spot Update Combining GIS Data and Remote Sensing Image

In order to improve the updating accuracy of class diagram spot, the paper puts forward a method of combination of GIS data and locally class diagram spot quick update in land Use status repository of high resolution remote sensing image. This method in premise of registration of remote sensing image and class diagram spot coordinate data, firstly uses the object-oriented image analysis method to classify the remote sensing image, and uses the existing class diagram spot data as prior knowledge of classification, in order to improve the accuracy of image classification, and then uses image classification results and class diagram spot data to judge the variation area, to update the variation area. The experimental results show that the proposed method can accurately update class diagram spot, and is a feasible method of land use status data update.