摘要
针对传统的水生植被遥感监测研究大多是面向大型浅水湖泊,利用Landsat和MODIS数据开展,且很少关注水生植被主要类群的细分,该文以小型湖泊-翠屏湖为例,利用欧空局Sentinel-2高分卫星数据,基于不同水生植被类群及水体间的光谱特征差异,构建了浮叶类植被指数(floating-leaved aquatic vegetation index,FAVI)和沉水植被指数(submerged aquatic vegetation index,SAVI)2个新的植被指数作为分类特征,结合Otsu算法,实现翠屏湖浮叶类植被类群、沉水植被类群和水体的自动提取。经验证,总体分类精度为88.57 %,Kappa系数为83.78%,并通过多期影像开展了算法的普适性检验。本研究为快速获取小型浅水湖泊的水生植被类群提供了高效的方法,可为湖泊管理和生态修复提供科学依据。
Traditional remote sensing monitoring of aquatic vegetation is mostly for large shallow lakes with Landsat and MODIS images and little attention is paid to the subdivision of major groups of aquatic vegetation.In this study,we develop two new spectral vegetation indices specifically targeted at aquatic vegetation,which are FAVI(floating-leaved aquatic vegetation index)and SAVI(submerged aquatic vegetation index)to realize the automatic extraction of floating-leaved aquatic vegetation group,submerged aquatic vegetation group and water body in Cuiping lake combined with Otsu algorithm,which is based on differences in spectral characteristics between different aquatic vegetation groups and water bodies with dense time series of high spatial resolution Sentinel-2 data from ESA.Based on validation,the overall classification accuracy is 88.57% and the Kappa coefficient is 83.78%.The universality test of the algorithm is carried out through multi-stage images.This study provides an efficient method for rapidly obtaining aquatic vegetation groups in small shallow lakes,providing a scientific basis for lake management and ecological restoration.
作者
汪政辉
辛存林
孙喆
罗菊花
马荣华
WANG Zhenghui;XIN Cunlin;SUN Zhe;LUO Juhua;MA Ronghua(Geography and Environment College of Northwest Normal University,Lanzhou 730070,China;Nanjing Institute of Geography and Limnology,Chinese Academy of Sciences,Nanjing 210008,China;Jiangsu Collaborative Innovation Center of Regional Modern Agriculture & Environmental Protection,Huaiyin Normal University,Huaian,Jiangsu 223001,China)
出处
《遥感信息》
CSCD
北大核心
2019年第5期132-141,共10页
Remote Sensing Information
基金
国家重点研发计划子课题（2016YFC0500201-05）
国家自然科学基金项目（41971314）
