基于高空间分辨率遥感影像的城市绿地提取方法研究

城市绿地是城市生态系统的一个重要子系统,在改善城市环境质量和居民生活水平方面扮演着越来越重要的作用。因此,如何及时获取城市绿地的分布及其变化,对于保护城市生态环境至关重要。本研究提出了一种基于高空间分辨率遥感影像的城市绿地提取方法,并珠海市为研究区域,利用Worldview-2遥感影像进行了城市绿地提取实验。结果表明,城市绿地提取结果比较理想,总体精度为94．09%,Kappa系数为0．94,但建设用地和裸地、城市绿地和建设用地之间仍然容易错分。

Extraction of Urban Green Space from High Spatial Resolution Remote Sensing Images

Urban green space is an important subsystem of the city ecosystem and plays a more and more important role in im-proving urban environment quality and resident living standards. Therefore, how to get the distribution and change of urban green space timely is vital to protect urban eco-environment. This research develops a method to extract urban green space information from high spatial resolution remote sensing images and takes Zhuhai City as a case study area to test the method using Worldview-2 images. Results show that classification result is ideal, the overall accuracy is 94. 09% and the Kappa coefficient is 0. 94, but there are still mistaken classification between construction land and bare land, between green space and construction land.