摘要
TM图像中由于裸地与城镇光谱特征相似,利用传统的分类方法难以区分二者,城镇提取精度很难令人满意.针对这一问题,本文提出了一种新的方法即比值居民地指数(RR I)法用于城镇信息提取,同时与最大似然监督分类法作对比,研究结果表明,RR I法(精度达87.50%)优于最大似然分类法(精度为78.13%),是一种提取城镇居民地信息的理想方法,尤其适合裸地较多的干旱半干旱地区.
In this paper, RRI （Ratio Resident-area Index） and Maximum Likelihood Classification （MLC） were used to retrieve urban residential areas in the region of Xi＇ an, respectively, from the satel- lite image of Landsat TM in 2003. Unlike conventional supervised classification for land use/cover retrieval, in this study, RRI can reflect the information of residential areas, and it is defined as RRI = TM1/ TM4. By comparing the two different methods, we find that the urban residential areas derived from TM imagery, using RRI is more accurate than that using MLC, the overall accuracy of them are 87.50% and 78. 13%, respectively. Results indicated that RRI is an effective way to retrieve urban residential areas. This method can not only obtain all the residential information, but also eliminate the influence of barrens, thus the retrieving accuracy is very high.
出处
《南京师大学报（自然科学版）》
CAS
CSCD
北大核心
2006年第3期118-121,共4页
Journal of Nanjing Normal University(Natural Science Edition)
基金
欧盟资助项目"SUSDEV-CHINA"(ICA4-CT-2002-10004)
中国科学院知识创新工程资助项目(KZCX3-SW-146)
作者简介
吴宏安，1981-，硕士研究生，主要从事遥感信息提取的学习与研究．E-mail：wha_105@yahoo．com．cn
通讯联系人：蒋建军，1963-，副教授，主要从事定量化遥感及其应用的教学与研究．E-mail：jiangjianjun@njnu．edu．cn
