基于ENVI的城市植被信息提取研究

Research on Urban Vegetation Information Extraction Based on ENVI

以陕西省西咸新区空港新城为研究区,采用2016、2018年夏季的Sentinel-2A遥感影像为数据,以ENVI为主要处理平台,分别采用面向对象和基于像元两种特征鲜明的图像分析技术对研究区影像提取植被信息,旨在分析两种方法对高分辨率遥感影像在城市植被信息提取中具有更好的分类效果,为基于遥感影像的城市地表信息提取提供借鉴。结果显示:面向对象法的分类精度为89.21%,而基于像元的分类精度为85.35%,面向对象自动分类算法能更好地提取城市植被信息。

With airport New City in Xi’ an-Xianyang New district as our research area in this study, and with the Sentinel-2 A remote sensing image in summer of 2016 and 2018 as the data, as well as the ENVI as the main processing platform, two image analysis techniques, object-oriented and pixels based image analysis technologies are respectively used to extract vegetation information from the image of the research area, aiming to analyze that the two methods are better for high-resolution remote sensing image in urban vegetation information extraction in terms of classification. The result of classification can be used for reference in urban surface information extraction based on remote sensing image. The results showed that the classification accuracy of the object-oriented method is 89.21%, while that based on pixel is 85.35%. The object-oriented automatic classification algorithm performs better in extracting urban vegetation information.