基于面向对象方法的露天煤矿用地类型提取优先级分析

Research on the Priority of the Land Use Types Extraction of Opencast Mine Area based on Object-oriented Classification

高分辨率影像为矿产资源开发遥感监管提供了更为精确有效的数据。以霍林河露天煤矿区为研究区,应用高分一号卫星影像为主要数据源,在面向对象的影像分类基础上,探讨了露天煤矿区用地类型信息提取优先顺序对最终分类精度的影响。结果表明:露天煤矿区的用地类型信息提取中,采用优先提取采矿场和排土场等资源开发用地类型、而后提取其他非开发用地的优先级顺序的分类精度最高,其总体精度达到82%,Kappa系数达到0.78,可以为露天煤矿区的用地类型信息提取提供理论和方法支持。

High-resolution images have provided more accurate data for the supervision of mineral resources exploitation.based on the image of satellite＂Gaofen-1＂,this test took Huolinhe open-cast coalmine as the study area,and adopted object-oriented classification method to discuss the influence of different classification orders on classification accuracy.Results indicated that in the land use types extraction of open-cast coalmine,the method of extracting stopes and refuse dumps fist had the highest classification accuracy,with the total accuracy to 82%and Kappa coefficient to 0.78.This method could offer theory and methodology to support for the land use types extraction of opencast mine area.