基于同日过空的Sentinel-2和Landsat 8数据的NDISI指数改进与应用  

Modification of NDISI by Using Sentinel-2 and Landsat 8 Data Acquired on the Same Day

作　　者：杨琳 丁凤[1,2,3] 朱伶俐 郑子铖 叶金玉 YANG Lin;DING Feng;ZHU Lingli;ZHENG Zicheng;YE Jinyu(Institute of Geography,Fujian Normal University,Fuzhou 35007,China;School of Geographical Sciences,Fujian Normal University,Fuzhou 350007,China;Fujian Provincial Engineering Research Center for Monitoring and Assessing Terrestrial Disasters,Fujian Normal University,Fuzhou 35007,China)

机构地区：[1]福建师范大学地理研究所,福建福州350007 [2]福建师范大学地理科学学院,福建福州350007 [3]福建省陆地灾害监测评估工程技术研究中心,福建福州350007

出　　处：《福建师范大学学报（自然科学版）》2020年第1期70-78,共9页Journal of Fujian Normal University：Natural Science Edition

基　　金：福建省自然科学基金资助项目(2017J01463);福建省科技厅公益类科研院所专项项目(2018R10344)

摘　　要：以福建省福州市为研究区,以同日过空的Sentinel-2和Landsat 8影像为主要数据源,采用MHUTS算法将Landsat 8热红外波段TIRS 10降尺度反演生成10 m空间分辨率的地表温度(LST)影像,然后将其引入基于Sentinel-2的10 m空间分辨率多光谱波段的NDISI指数的构建,提出了一种新型的不透水面指数MNDISIS2-L8,并对改进前后的研究区不透水面提取结果进行了精度验证和比较分析.结果表明:经MHUTS算法空间降尺度后,LST影像的空间分辨率得到了明显的提高,细节得以清晰展示的同时也较好地保持了原始影像的热特征.改进前,研究区不透水面提取总精度和Kappa系数分别为88.60%和0.730;改进后,不透水面提取总精度达94.75%,Kappa系数为0.881.In this paper,a new index,modified NDISI based on Sentinel-2 and Landsat 8 images(MNDISIS2-L8),was put forward for estimating impervious surface.Taking Fuzhou,the capital city of Fujian Province,China,as the study area,two date-coincident images of Sentinel-2 and Landsat 8 acquired on 16 September,2017 were used.Due to the lack of a thermal band in Sentinel-2 imagery,the MHUTS method was introduced to obtain fine-sharpened 10 m resolution land surface temperature(LST) image from Landsat 8 TIRS 10,and after that,the LST image was added to the Sentinel-2 10 m multi-spectral bands to calculate MNDISIS2-L8.Comparisons of the performances of the original NDISI(by using Landsat 8 data alone) with the MNDISIS2-L8 were made.The results showed that,by introducing MHUTS,the spatial resolution of the LST image was significantly improved and the details were clearly displayed while maintaining the thermal characteristics of the original image.With the original NDISI,the overall accuracy and Kappa coefficient of the impervious surface extraction in the study area were 88.60% and 0.730,respectively.With the new index,the overall accuracy was 94.75%,and the Kappa coefficient was 0.881.

关 键 词：Sentinel-2 Landsat 8 MHUTS NDISI MNDISIS2-L8 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]