基于高分辨率遥感影像的土地覆盖信息提取

Extracting Land Cover with High Resolution Remote Sensing Data

高空间分辨率遥感影像使得土地覆盖和土地利用信息的提取成为可能。以高分辨率遥感影像数据IKONOS为主要数据源,以多尺度分割与基于模糊逻辑分类的面向对象影像分析方法为主要技术,自动提取株洲市城乡结合部的土地覆盖和土地利用信息。达到了提取郊区丘陵地带林地信息和城市建筑、道路等土地覆盖信息的目的,而且精度高,速度快。结果表明利用该方法对复杂的城乡结合部信息获取是可行的。

As the main data sources for land cover and land use, high-resolution imagery provides a good basis for extracting land cover and land use information. Going far beyond the methodical limits ot pixelbased and manual interpretation approaches, multi-scale image segmentation , classification based fuzzy logic and objcot-oriented image analysis approaches are used for extracting information from remote sensing data. This paper presents a snapshot of work to extracting information in Zhuzhou city between country and urban. It allows the segmentation of an image into highly homogeneous image objects in any chosen scale and the generation of a network of image objects. The process does not classify single pixel but rather image object. Not only spectral information but also spatial, physical and contextual characteristics of image objects are used for classification. Classification is conducted by fuzzy logic. The result of land cover information extraction is promising and the precision of classification is wonderful. It is obvious that such new image analysis approach offers a satisfying solution to extract information quickly and efficiently.