一种基于TVDI模型的边界提取方法研究

浏览次数：14

分享到:

摘要:

提出一种基于TVDI模型的湿地边界信息提取方法,利用TM遥感影像建立地表温度(Ts)与归一化植被指数(NDVI)之间的二维特征空间,用IDL编程在特征空间内拟合温度植被干旱指数法(TVDI)需要的干、湿边方程,并反演出试验区的土壤干湿情况.利用野鸭湖湿地3m×3m植物样方,统计、分析、确定湿地生态系统的上边界.试验测得湿地边界的土壤水分含量阈值0.56,根据该阈值确定湿地边界信息,提取湿地边界.通过该方法提取的湿地核心区准确率达到72%,说明利用遥感方法反演土壤水分含量,提取湿地边界的方法可行.

The extraction of soil moisture of wetland is of great significance to the wetland system.So this paper presents a model based on the ideal of this extraction to get the wetland boundary.The two dimensional feature spaces of surface temperatures（Ts）and normalized differential vegetation index（NDVI）is established by using the TM remote sensing images,and the dry and wet side equation which is necessary in the TVDI model is fitted in the feature space with the programming of IDL.Using observed data of plant sample from the wild to determine the reference upper boundary.After statistics and analysis,the threshold of soil moisture is identified as 0.56.The core area of wetlands accuracy of the method of extraction by 72%,indicating that the inversion of soil moisture through remote sensing methods to extract the wetland boundary is feasible.

作者:

阿多赵文吉宫兆宁周廷刚

机构地区:

首都师范大学资源环境与旅游学院/三维信息获取与应用教育部重点实验室/资源环境与地理信息系统北京市重点实验室西南大学

出处:

《Betway官方客服学报：自然科学版》 CAS 北大核心 2015年第2期158-163,共6页

基金:

国家国际科技合作专项(2014DFA21620) 国家留学基金项目

关键词:

温度植被干旱指数法湿地边界信息提取

TVDI soil moisture inversion wetland boundary extraction

分类号: