摘要
遥感影像分类在专题信息提取、地表动态监测以及专题地图制作等应用中具有重要作用,传统的分类方法可以分为监督分类和非监督分类,因算法成熟、操作简单,这两类方法仍然是当前使用较广泛的分类方法,但从理论、过程以及使用范围条件上二者都不相同,各有其优缺点。鉴于这种现状,本文采用Landsat-8 OLI焦作地区遥感数据分别基于监督与非监督中的各种算法进行土地覆盖分类,并对分类结果进行比较分析和精度评价,以期为实际工作中根据不同需求选取适当分类器提供依据。研究结果表明:监督分类中最大似然法分类精度相对较高,漏分错分最少,总体分类精度达到87.152%;非监督分类中ISODATA算法从聚类效果、漏分错分以及计算时间上综合分析要优于K-均值分类;另外,不同分类算法对不同地物类型的解译效果不同。
Remote sensing image classification plays an important role in the thematic information extraction,dynamic monitoring of surface and making thematic map,The traditional classification methods can be divided into supervised and unsupervised classification,Because of its mature algorithm and simple operation,But there are some different in theory,process and the range of application,Each has its advantages and disadvantages.In view of this situation,This paper uses the Landsat8 OLI of Jiaozuo area to classify the land cover based on the supervised and unsupervised classification in various algorithms respectively,And the classification results are comparative analysis and accuracy evaluation,In order to provide the basis for practical work according to the different needs to select the appropriate classifier.The research results show that the classification accuracy of the maximum likelihood supervised classification is relatively high,The leakage error is the least and the overall classification accuracy can be up to 87.152%;The ISODATA algorithm is better than the K-means in unsupervised classification from comprehensive analysis of the clustering effect,leakage error and calculation time;In addition,Different classification algorithms have different interpretation effects for different ground features.
作者
张明
黄双燕
ZHANG Ming;HUANG Shuangyan(Department of Surveying and Mapping Engineering,Northeastern University,Shenyang 110819,China;Xinjiang Institute of Ecology and Geography,Chinese Academy of Sciences,Urumqi 830011,China;University of Chinese Academy of Sciences,Beijing 100049,China)
出处
《测绘与空间地理信息》
2019年第1期177-180,共4页
Geomatics & Spatial Information Technology
关键词
焦作市
监督分类
非监督分类
精度评价
Jiaozuo City
supervised classification
unsupervised classification
accuracy evaluation