摘要
根据不同地物类型在遥感影像上的光谱、纹理、几何和位置响应特征的差异性,可以对地物进行分类。本文采用面向对象的方法,基于影像上地物类型的纹理和光谱等特征,利用ENVI EX面向对象特征提取模块对试验路域SPOT5影像进行地物分类,从而定量动态分析地物类型变化情况;并通过提取NDVI对植被覆盖度进行了分级统计分析。结果表明,定量分析成果与野外调研实验结果相吻合,可以用来评价实地植被覆盖度及地物分类变化情况,进而评价公路路域植被环境,为公路的可持续发展提供基础支持。
Based on the different cover types in the remote sensing image with the different spectrum,texture,geometry and location.In this paper,depending on the texture and spectral features of the image,we used the object-oriented classification method to do the classification on the SPOT5 test road domain image by using the ENVI EX object oriented feature module.Then we can do quantitative analysis of the changes on the objects,and by extracting the NDVI to do classification statistics of the vegetation coverage.The results showed that,the quantitative analysis results were in good agreement with the field investigation results,that can be used to evaluate the vegetation coverage and the classification changes situation.Then the evaluation of the vegetation environment of highway domain could be done,which would provide the theoretical support for the sustainable development of the highway.
出处
《北京测绘》
2016年第1期39-42,80,共5页
Beijing Surveying and Mapping
基金
国家自然科学基金面上项目(41471421)
湖南省科技项目计划(2015WK3022)
关键词
面向对象
路域
植被环境
分类
NDVI
object-oriented
road domain
vegetation environment
classification
NDVI
