Abstract: Soil moisture is a crucial parameter for the exchange of matter and energy at the land-atmosphere interface. Timely and accurate acquisition of soil moisture information is of paramount importance for drought monitoring, water resource management, and crop yield estimation. In this study, utilizing Sentinel-1 SAR remote sensing data and Sentinel-2 optical remote sensing data, the relationship between various optical vegetation indices and measured vegetation water content was systematically analyzed. The Fusion Vegetation Index（FVI） was preferentially selected to establish vegetation water content estimation model,which was combined with the vegetation microwave scattering model—Water Cloud Model（WCM） to correct the impact of vegetation layer on SAR backscattering signals. On this basis, a surface microwave scattering model—Oh model was used to construct the backscattering coefficient simulation database, and soil moisture retrieval for the summer maize-covered surface under both VV and VH polarizations was achieved through the application of the Look-Up Table（LUT） algorithm. The results indicate that, for surfaces covered by dense vegetation like summer maize, vegetation water content characteristics can be better reflected by FVI, enabling the accurate correction of the impact of vegetation layers on SAR backscattering coefficients. The vegetation water content inversion model based on FVI achieved an R2 of 0.693 and an RMSE of 0.303 kg/m2. After vegetation correction, the correlation between soil moisture and SAR backscattering coefficients increased by 21.6% and 27.9% for VV and VH polarizations, respectively. Compared to VH polarization, VV polarization is found to be more suitable for soil moisture retrieval, with an R2 of 0.672 and an RMSE of 0.048m3/m3 between retrieved and measured soil moisture values. The findings of this study provide robust support for the remote sensing observation of soil moisture information in densely vegetated surfaces.