Abstract: Remote sensing images can acquire the spectrum, tone, texture and other information of the surface rocks, but the extracted information is limited to the surface and has no obvious advantage in the interpretation of deep geological information.Aero geophysical data has more advantages in extracting information of underground abnormal bodies. A single type of data cannot meet the needs of complex applications in basic geology and resource exploration.Therefore, a combined method of remote sensing and aero geophysical analysis was proposed. A certain area in Xinjiang was taken as the study area to classify the lithology through the analysis of multi-source remote sensing images and aero geophysical data based on the random forest method.The results show that the proposed method can improve the accuracy of lithologic classification in the study area compared with a single type of data.The proposed method has certain practical value and guiding significance in promoting the fine application of remote sensing and aero geophysical exploration technology in geological mapping.