Soil moisture is the link between the surface and atmospheric circulation. It is an important part of hydrological cycle and water-heat balance. And it plays a key role in vegetation growth and efficiently agricultural irrigation. Therefore, accurate and fast soil moisture remote sensing inversion results are importantly basic data for agricultural production and ecological environment protection. The MODIS-MYD09A1 data with a spatial resolution of 500 meters and a temporal resolution of 8 days was used to calculate the surface albedo. The MODIS-MYD13A1 NDVI product data with a spatial resolution of 250 meters and a temporal resolution of 16 days was used to calculate vegetation coverage. In order to improve the estimation accuracy of the remote sensing soil moisture in the Shiyang River Basin, bare soil albedo was calculated by using vegetation coverage and surface albedo trapezoidal characteristics space scatter plot, this is to reduce the error of vegetation on soil moisture in remote sensing. The measured data was used to evaluate the accuracy of the inversion model. The inversion results of the bare soil albedo model were used to grade and map. At the same time, the spatial pattern of SM and its influencing factors were analyzed through stability analysis, spatial autocorrelation analysis and geographic detectors. The results show that: (1)The bare soil albedo model could yield more accurate soil moisture inversion accuracy in the Shiyang River Basin, which provided a new method for the SM calculation at the watershed scale. (2)Soil moisture had obviously spatial autocorrelation, Moran's value was 0.88 (Z-score=1852.94, P<0.01). Forest land in the upstream was high-high aggregation. Desert in the downstream was low-low aggregation. Soil moisture was significantly correlated with fractional vegetation coverage (P<0.01). (3) The annual overall stability of the soil moisture in the Shiyang River Basin being good, good stability and better stability area accounted for 88.34% of the study area. (4) The spatial distribution of soil moisture was affected by multifactor. And the explanatory ability of each factor was significantly different. Among these factors, vegetation coverage had the best interpretive ability. Soil type was in the second place. Elevation was in the third place and land use was at the end.The interaction between factors enhanced the explanatory ability of the spatial differentiation of soil moisture. (5) The soil moisture of different land use types was quite different. Among them, the soil moisture of most unused land was less than 7%; the soil moisture of grassland and cultivated land was at medium level, with soil moisture value of 7%-15%; the soil moisture level of woodland was the highest, and soil moisture value more than 25%.