Abstract: In order to quickly and accurately monitor the coverage of floating leaf vegetation, this article focuses on the sub-lakes and isolated lakes of Poyang Lake which with widespread floating leaf vegetation. First, the Normalized Difference Vegetation Index (NDVI) of the pixels were calculated based on Sentinel-2satellite imagery. Then, corresponding field coverage of floating leaf vegetation (FCFLV) in the pixels were assessed using drone aerial imagery, and a regression model between NDVI and FCFLV is established. Finally, using Dawu Lake as an example, floating leaf vegetation’s coverage (FLVC) of Dawu Lake was inverted using both the method presented in this paper and the traditional pixel dichotomy method, and the inversion accuracy of the two methods was compared. The results indicate: 1) The regression model based on the NDVI and FCFLV demonstrates a very good fit, with a coefficient of determination (R²) reaching 0.9; the model's root mean square error (RMSE) is 5.75%, and the mean relative error (MRE) is only 9.0%; 2) Using the traditional threshold dichotomy method to invert FLVC yields a minimum RMSE of 20.25% and a minimum MRE of 53.68% when the NDVI threshold is 0.081, both of which are significantly higher than the RMSE and MRE obtained from the method in this paper; 3) Compared to the traditional pixel dichotomy method, this model can more accurately invert FLVC, especially in areas where the vegetation is sparse. The method presented in this paper, which constructs a regression model based on drone aerial data and satellite NDVI, can provide a reference for the rapid monitoring and quantitative inversion of aquatic plant communities.