Analysis of Landsat-8 OLI imagery for land surface water mapping
Analysis of Landsat-8 OLI imagery for land surface water mapping
复制标题
用于陆地表面水绘图的 Landsat-8 OLI 图像分析
DOI：
10.1080/2150704x.2014.960606
复制
发表时间：
2014-01-01
影响因子：
2.3
通讯作者：
Sun, Bingyu
中科院分区：
文献类型：
--
作者：
Du, Zhiqiang;Li, Wenbo;Sun, Bingyu
将归一化差分水指数(NDWI)成功地应用于陆地卫星系列多光谱图像的陆地表层水制图(LSWM)。本文利用三个NDWI模型评估了最近的陆地卫星(Landsat-8)运行的陆地成像仪(OLI)多光谱图像用于LSWM的潜力。我们在中国对长江流域和淮河流域的三个OLI NDWI模型的精度和稳健性进行了检验。结果表明，三个OLi NDWI模型对LSWM的总体准确率在95%以上，卡帕系数为0.89，生产者准确率为95%。结果还表明，基于OLI传感器的绿波段(Band 3)和SWIR1波段(Band 6，3)的NDWI模型比其他两个NDWI模型具有更高的LSWM精度。
The normalized difference water indices (NDWIs) were successfully used in map land surface water mapping (LSWM) from Landsat series multispectral images. This paper evaluates the potential of the recent Landsat satellite (Landsat-8) Operational Land Imager (OLI) multispectral images for LSWM using three NDWI models. We tested the accuracy and robustness of the three OLI NDWI models in the Yangtze River Basin and the Huaihe River Basin in China. The results demonstrate that the three OLI NDWI models achieve an overall accuracy of more than 95%, a kappa coefficient of 0.89 and a producer’s accuracy of 95% for LSWM. The results also demonstrate that the NDWI model using the green band (Band 3) and the SWIR1 band (Band 6) (referred to as NDWIO6,3) of the OLI sensor has a higher LSWM accuracy than the other two NDWI models.