一种区分平原区城镇与农村用地新方法

摘要

区分城镇与农村用地是遥感专题信息提取的关键.针对遥感影像很难有效区分城镇与农村用地的问题,结合稳定夜间灯光影像和增强型植被指数影像,构建区分平原区城镇与农村用地新方法,对南京城镇与农村用地提取.利用人工解译结果对研究区提取结果进行精度评价,并对比不同城市进行精度验证.结果表明,该方法对南京居民地提取总体精度和Kappa系数分别为86.89％和0.64.其中,南京城镇用地提取总体精度和Kappa系数分别为91.38％和0.66,农村用地精度分别为90.31％和0.35.不同城市提取城镇与农村总体精度分别达到85.34％和80.27％,提取结果与人工解译结果空间格局基本吻合.该方法从居民地中区分城镇与农村,是一种有效的辨别城镇与农村用地遥感提取方法,为区域城镇与农村用地信息区分提供新思路.%It is the crucial part of remote sensing thematic information extraction to distinguish urban and rural land information.It is difficult to distinguish urban and rural land from remote sensing images.Based on DMSP/OLS stable nighttime light images and MODIS-NDVI images,this paper put forward a new method to extract urban and rural land in plain terrain in Nanjing.The result of artificial interpretation was used to evaluate the accuracy of the extraction of the study area,and the precision verification of different cities was compared.The results showed that the overall accuracy and Kappa coefficient of the residents were 86.89％ and 0.64 in Nanjing.The overall accuracy and Kappa coefficient of urban land in Nanjing were 91.38％ and 0.66,respectively.Meanwhile,the overall accuracy and Kappa coefficient of rural land were 90.31 ％ and 0.35.The overall accuracy of urban and rural land in different cities reached 85.34 ％ and 80.27 ％.The results of the new method extracted were in accordance with the spatial pattern of the artificial interpretation results.The method is an effective method to distinguish urban and rural land using remote sensing,which puts forward a new idea to distinguish region urban and rural land information.