Atmospheric aerosol is an important influence factor of climate change and remote sensing quantification, so it is of great significance to study aerosol optical properties. Based on the measurements of directional polarimetric camera (DPC) onboard GF-5 satellite over China, the retrieval of aerosol optical depth (AOD) is carried out. Based on the vector radiative transfer model 6sv, a look-up table of aerosol optical properties is constructed. Then the contribution of land surface is calculated by Ross Li surface BRDF model and semi empirical NB model. Finally, AOD is retrieved by using multi angle information of intensity and polarization and look-up table method. It is found that the retrieval results of DPC aerosol optical depth are in good agreement with the observation results of AERONET ground stations in China, the linear fitting degree is good and the correlation coefficient is greater than 0.8, which verifies the reliability of the algorithm.And it is believed that the development of the algorithm provides technical support for DPC to effectively monitor the spatial and temporal distribution of aerosols.