基于 GEE 平台的黄河流域水体指数研究.

The water body in the Yellow River basin is complex. Rapid and accurate water extraction in the basin is of great significance to the ecological protection and high-quality development of the Yellow River Basin. However, the existing water body indexes can not meet the requirements of removing the inherent mountain shadow of remote sensing images and accurately extracting complex water bodies. Based on Google Earth Engine (GEE) platform and Sentinel-2 remote sensing images, this paper proposed a Yellow River Basin Water Index (YRBWI), and the effect of water extraction results were compared with the Normalized Difference Water Index (NDWI), Modified Normalized Difference Water Index (MNDWI) and Shadow Water Index (SWI). The results show that YRBWI achieves an excellent water extraction effect in the Yellow River Basin, specifically: YRBWI can solve the confusion of mountain shadows and accurately extract water bodies with high sediment content; compared with NDWI, MNDWI, and SWI, YRBWI has the largest OA and Kappa values, indicating that YRBWI has the highest water extraction accuracy; the water body extraction accuracy of YRBWI is insensitive to the change of threshold value and has high robustness. [ABSTRACT FROM AUTHOR]