An absolute radiometric calibration method based on the surface hyperspectral Bidirectional Reflectance Distribution Function (BRDF) model is proposed, which the efficiency and frequency of on-orbit absolute radiometric calibration of remote sensors are improved by getting rid of the dependence on synchronous observation of satellite overpass time. In-flight high-frequency absolute radiometric calibration verification and application was carried out with Operational Land Imager (OLI) on Landsat-8 using this method. The surface directional reflectance of the Dunhuang test site was measured by the Unmanned Aerial Vehicle (UAV) measurement system, and the hyperspectral BRDF model parameters of the Dunhuang site were inversed based on the semi-empirical kernel-driven model. Comparing the model calculated data with the measured data, the present model is proven to provide a better fit to measured data. The surface hyperspectral reflectance data of satellite observation direction calculated directly by surface BRDF model is used to replace the synchronous measurement data of satellite transit time. The high-frequency absolute radiometric calibration of Landsat-8 OLI is realized by using the reflectance basis method combined with MODIS official atmospheric data products. The relative deviation between the Landsat-8/OLI observed apparent radiance and the model calculated apparent radiance is less than 5% and the standard deviation is less than 2% in the first band to the sixth band. The result shows that the absolute radiometric calibration method based on the UAV BRDF model are in high consistency and stability with the satellite observations.