【摘要】 富贵竹作为一种观赏植物,在我国南方省份有大面积种植,具有良好的经济价值。为了解与监测区域富贵竹种植情况,以Landsat8 OLI遥感影像为数据源,构建了一种新的网体—水体差异增强指数(difference enhence between net and water index,DENWI)作为特征参数,通过面向对象的分类方法建立富贵竹信息提取规则集,得到研究区域内富贵竹的种植信息,并与2种传统信息提取方法进行对比研究。结果表明,相比于传统方法,基于DENWI的面向对象分类方法可以更有效地提取富贵竹种植信息,总体分类精度为98. 46%,Kappa系数为0. 97,该方法监测提取富贵竹种植信息是可行且有优势的,可以为富贵竹种植监测和管理提供科学依据。更多还原

【Abstract】 The Dracaena sanderiana,as an ornamental plant,has been extensively planted in southern China,and has good economic value.In order to monitor the planting situation of Dracaena sanderiana,the authors constructed a new index-"difference enhence between net and water index"(DENWI) as a characteristic parameter based on Landsat8 OLI remote sensing image.Object-oriented classification method was used to establish the Dracaena sanderiana information extraction rule set,the Dracaena sanderiana planting information was obtained,and two kinds of traditional information extraction methods were adopted for comparative study.The results show that,compared with the traditional method,the object-oriented classification method based on DENWI can extract the information of Dracaena sanderiana,with the overall classification accuracy being 98.46% and the kappa coefficient being 0.97.Remote sensing monitoring and extraction of Dracaena sanderiana planting information is feasible and advantageous,and it can provide scientific basis for monitoring and management of Dracaena sanderiana.更多还原