摘要
本文在分析现有居民地提取方法的基础上,提出将归一化建筑指数(NDBI)、改进归一化差异水体指数(MNDWI)、土壤调节植被指数(SAVI)、比值居民地指数(RRI)相结合进行居民地信息提取的方法。以浙江省宁波市为例,通过光谱采样及各类地物在4种指数上的取值分析,建立模型进行居民地信息提取及精度验证,结果表明:利用该模型可以实现居民地信息的自动提取,能提高居民地与裸地的可分性,减少背景地物的影响,总体精度为91.08%。
In order to solve the problem of how to extract residential areas from background features, especially in distinguishing bare land and residential areas, this paper analyzed each method of extracting residential areas from TM images, proposed combination of spectral characteristics and Normalized Difference Built-up Index （NDBI） , Modified Normalized Difference Water Index （MNDWI） , Soil Adjusted Vegetation Index （SAVI） and the Ratio of Residents to Index （RRI） to extract residential areas. Taking Ningbo city as a case area, it established an extraction model based on a variety of indexes. Experimental results showed that the model could realize residential areas automatic extraction form TM images, and improve the divisibility between residential area and the bare ground, and the accuracy of extraction was 91.08%.
出处
《测绘科学》
CSCD
北大核心
2013年第2期146-149,共4页
Science of Surveying and Mapping
基金
国家科技支撑计划项目(2008BAK50B04)
科技部国际合作项目(2009DFR20620)
国家公益性行业(气象)专项(GYHY201006039)
