Abstract: Loess landslide disasters occur frequently and are widely distributed in China. Traditional geological hazard surveys are difficult to effectively identify hidden landslide hazards that are located at high altitudes, have unclear deformation characteristics, and are hidden. This is also one of the main reasons for the low success rate of landslide hazard monitoring and warning. How to effectively identify geological hazard hazards beyond prejudgment is the premise and foundation of geological hazard prevention and control work. Time series InSAR technology has good application potential in this field, but how to better integrate InSAR technology into landslide disaster related research is still in the exploratory stage. The author takes the southwest area of Bailuyuan in Xi’an City as the research area, and on the basis of high−precision 3D oblique photography, ALOS-2 radar image set, and other data, uses time-series InSAR technology to invert 104 areas with obvious surface deformation. By combining the susceptibility index of loess landslides, aerial images, and field verification, 23 loess landslides and hidden dangers were quickly identified, including 20 newly identified landslide hazards and 3 registered landslide disasters. The advantages and effectiveness of the time−series InSAR method detection results were verified through comparison with traditional geological disaster investigation data and on−site investigation verification. A high−precision InSAR and DEM data based early identification method for loess landslide hazards was constructed.