文献类型：期刊文献

中文题名：基于改进的面向对象遥感影像分类方法研究——以西藏米林县典型林区为例

英文题名：Optimized Methodology for Classification of Remote Sensing Images based on Object-oriented:Application to the Typical Forest of Milin in Tibet

作者：施佩荣[1] 陈永富[1] 刘华[1] 吴云华[2] 魏新[2] 钟泽兵[2]

第一作者：施佩荣

机构：[1]中国林业科学研究院资源信息研究所;[2]西藏自治区林业调查规划研究院

年份：2017

卷号：32

期号：3

起止页码：466-474

中文期刊名：遥感技术与应用

外文期刊名：Remote Sensing Technology and Application

收录：CSTPCD;;北大核心:【北大核心2014】；CSCD:【CSCD2017_2018】；

基金：国家科技基础性工作专项(2013FY111600)"中国森林植被调查"

语种：中文

中文关键词：西藏;面向对象;最优分割尺度;最近邻法;阈值分类法

外文关键词：Tibet; Object-oriented; Optimal segmentation scale; Nearest neighbor classification; Threshold classification;

分类号：TP79

摘要：针对目前面向对象多尺度分类方法中,最优分割尺度的确定方法不具有普适性或者易受主观性影响的问题,以西藏米林县的Landsat 8OLI影像为数据源,对研究区影像多尺度分类进行研究。首先确定多尺度分类的最优分割尺度,提出基于多尺度分类精度Kappa系数的最优分割尺度函数模型法,在此基础上,利用多尺度分类分别与最近邻分类和阈值分类法相结合的方法,对研究区影像进行分类。结果表明:分割尺度分别为190、150、100、60,多尺度分类法比单一尺度分类精度高;最近邻多尺度分类法比阈值多尺度分类精度高,其总精度分别为86%和72%,Kappa系数分别为0.72和0.69。最优分割尺度函数模型在具有普适性的基础上更具有科学理论性,多尺度分类与最邻近分类结合的方法比与阈值分类结合的方法分类效果好,为后续植被动态变化监测提供了依据。
For the problem of limitation of the mainly subjective or lacking of widely used of the methods of mult-scale object-oriented classification,the method of multi-scale classification for the research.The data source was Landsat-8OLI images of Milin county in Tibet.First,the optimal segmentations of multi-scale segmentation should be confirmed.This paper proposed the function model between the accuracy of multiscale classification and segmentation scales.Then,classified the images with the methods of multi-scales combined with nearest neighbor classification and threshold classification respectively.The result showed that the segmentation scales was 190,150,100,60,and the accuracy of multi-scale segmentation is higher than the single one.Others,the accuracy of nearest neighbor classification with the multi-scale is higher than the threshold classification＇s,the accuracy was 86% and 72%,and the kappa was 0.72 and 0.69.The function model of optimal segmentation scale is a method can be used widely with scientific theoretical.The method of nearest neighbor classification and threshold classification with multi-scale provide a foundation for the change of forest.