基于决策树方法的Landsat8 OLI影像红树林信息自动提取
发布时间:2017-12-26 17:25
本文关键词:基于决策树方法的Landsat8 OLI影像红树林信息自动提取 出处:《国土资源遥感》2016年02期 论文类型:期刊论文
更多相关文章: 红树林 NDMI MNDPI OLI 决策树
【摘要】:基于广西山口国家红树林生态自然保护区的Landsat 8 OLI影像数据,选用广泛应用于植被液态水含量反演的归一化差值湿度指数(normalized difference moisture index,NDMI)和修正的归一化差值池塘指数(modified normalized difference pond index,MNDPI)作为分类特征,运用决策树方法进行红树林信息的自动提取。研究结果表明:红树林独特的滨海湿地生境特点,使其光谱同时包含植被和湿地信息;MNDPI和NDMI可分别反映可见光-近红外波段反射率同短波红外波段反射光谱的反差,可成功应用于湿地植被信息的提取,能有效地将红树林同其他地物相区分;采用Landsat8 OLI遥感数据,并结合NDMI和MNDPI分类特征构建的决策树模型可有效地提取红树林信息,其错分率和漏分率都较低,分别为5.34%和1.69%。
[Abstract]:Guangxi Mountain National Mangrove Nature Reserve of Landsat 8 OLI based on the image data, the widely used in the normalized difference vegetation index and humidity inversion of liquid water content (Normalized Difference Moisture index, NDMI) and modified normalized difference index (modified normalized difference pond the index, MNDPI) as the classification feature extraction using decision tree method of mangrove information. The results showed that the mangrove coastal wetland habitat features unique, the spectrum contains both vegetation and wetland information; MNDPI and NDMI represent contrast visible near infrared reflectance with SWIR reflectance spectra, can be successfully applied to the extraction of wetland vegetation information, can effectively combine with other objects of mangrove distinguish; using Landsat8 OLI remote sensing data, and combined with the decision tree model to construct the NDMI and MNDPI features can effectively extract the mangrove information, error rate and leakage rate are low, 5.34% and 1.69% respectively.
【作者单位】: 南京信息工程大学地理与遥感学院;
【基金】:国家自然科学基金项目“红树林冠层高光谱探测及群落类型识别研究”(编号:41201461) 江苏政府留学奖学金共同资助
【分类号】:TP751
【正文快照】: 0引言红树林是生长在热带、亚热带沿海潮间带滩涂上特有的木本植物群落,属常绿阔叶林,主要分布于淤泥深厚的海湾或河口盐渍土壤上。红树林具有促淤固滩、防浪护堤、保持生物多样性和净化环境污染等作用[1-2]。然而由于红树林生存于海洋与陆地交错的生态脆弱带,长期受到沿海不,
本文编号:1338169
本文链接:https://www.wllwen.com/guanlilunwen/gongchengguanli/1338169.html