摘要
针对MODIS(moderate resolution imaging spectroradiometer,中分辨率成像光谱仪)影像浒苔提取中存在的植被指数自适应阈值不普适等问题,本文提出了多指数决策融合的MODIS浒苔提取算法(multi-index decision fusion,MIDF)。计算影像的归一化植被指数(normalized difference vegetation index,NDVI)、增强型植被指数(enhanced vegetation index,EVI)、差值植被指数(difference vegetation index,DVI)和比值植被指数(ratio vegetation index,RVI),采用局部阈值法对4个指数灰度图进行自适应阈值分割得到初始浒苔覆盖范围,通过“绝对多数投票法”对上述初始浒苔覆盖范围投票表决得到最终提取结果。实验结果表明,MIDF可适应不同密度的浒苔覆盖区域,且精度优于传统的NDVI、EVI、DVI和RVI自适应阈值法。本算法为无监督法,自动化程度较高,可应用于浒苔灾害遥感业务化监测。
With the aim at solving the problem that adaptive threshold of vegetation index is not universal in the extraction of Enteromorpha prolifera from moderate resolution imaging spectroradiometer(MODIS)images,this paper proposed a multi-index decision fusion(MIDF)algorithm for Enteromorpha prolifera extraction.Firstly,the normalized difference vegetation index(NDVI),enhanced vegetation index(EVI),difference vegetation index(DVI)and ratio vegetation index(RVI)of the image were calculated,and then the local threshold method was used to segment the four gray-scale images of vegetation index to obtain the four initial coverage areas of Enteromorpha prolifera.Finally,the"absolute majority voting method"was used to vote on the four initial coverage areas of Enteromorpha prolifera,and the final result of Enteromorpha extraction was obtained.The experimental results showed that MIDF can adapt to the coverage areas of Enteromorpha with different densities,and performaned better than traditional NDVI,EVI,DVI and RVI adaptive threshold methods.MIDF is an unsupervised algorithm with a high level of automation,and can be applied to remote sensing operational monitoring of Enteromorpha disasters.
作者
万剑华
万献慈
刘善伟
盛辉
许明明
WAN Jian-hua;WAN Xian-ci;LIU Shan-wei;SHENG Hui;XU Ming-ming(College of Oceanography and Space Informatics,China University of Petroleum,Qingdao 266580,China)
出处
《海洋环境科学》
CAS
CSCD
北大核心
2022年第2期316-324,共9页
Marine Environmental Science
基金
国家自然科学基金(41776182)
山东省自然科学基金(ZR2019MD023)。
关键词
MODIS
浒苔
多指数
自适应阈值
决策融合
MODIS
Enteromorpha
multi-index
vegetation index adaptive threshold
decision fusion
