遥感图像监督分类与面向对象分类法对比研究

发布时间：2018-01-06 23:14

本文关键词：遥感图像监督分类与面向对象分类法对比研究　出处：《农村经济与科技》2016年23期 　论文类型：期刊论文

【摘要】：最近几年,随着空间遥感技术的发展,高分辨率遥感影像在土地资源调查、环境监测等领域中的应用日益广泛。利用传统的遥感影像分类方法,如监督分类,容易导致分类精度降低,空间数据大量冗余和资源的浪费。面向对象的分类方法正是为了解决这些问题而出现,面向对象的方法利用遥感影像结构、光谱信息,并在建立这些特征之间的层次关系的基础上,对遥感影像进行分类。利用杭州市西湖区域2014年10月份的数据,基于ENVI5.0对高分辨率影像进行地物特征的信息提取,利用监督分类和面向对象技术首先对该区域信息进行分类,然后再对分类结果进行对比分析研究。结果表明,面向对象分类方法在此数据城市绿地信息提取中精度优于监督分类。
[Abstract]:In recent years, with the development of space remote sensing technology, high-resolution remote sensing images in land resources survey, environmental monitoring and other fields are increasingly widely used, using traditional remote sensing image classification methods, such as supervised classification. It is easy to reduce classification accuracy, redundancy of spatial data and waste of resources. Object-Oriented classification method is to solve these problems. Object-Oriented method uses remote sensing image structure and spectral information. On the basis of establishing the hierarchical relationship between these features, the remote sensing images are classified. The data of October 2014 in the West Lake region of Hangzhou City are used. Based on ENVI5.0, the feature information of high resolution image is extracted, and the region information is classified by supervised classification and object oriented technology. Then the classification results are compared and analyzed. The results show that the accuracy of the object-oriented classification method is better than that of supervised classification in the data extraction of urban green space information.
【作者单位】：贵州大学资源与环境工程学院;贵州大学矿业学院;
【分类号】：P237
【正文快照】： 监督分类是以像元为基本单元进行遥感信息提取的分类方法,常用的监督分类有最大似然法、最短距离法、马式距离法等常用的方法。它主要根据像元的光谱信息进行分类,然而,高分辨率遥感数据通常包含较少波段,光谱信息不如空间特征丰富,分类时不能仅靠光谱特征,更多的是要利用其几

【相似文献】