Abstract: Ground-based synthetic aperture radar (SAR) has the technical advantages of non-contact, high-precision, large-area, all-weather all-weather, 24-hour continuous monitoring. It was an important technical means for monitoring and early warning of landslide disasters. The detection of permanent scatterers was the key technology for deformation inversion. However, when the permanent scatterer is screened by the threshold method, the manual debugging of multiple thresholds was complicated and the threshold setting is unreasonable. In this paper, a multi-threshold optimization algorithm was proposed, which automatically determines and determines the threshold of permanent scatterers (PS) points for ground-based radar, which can reduce the irrational phenomenon of manually setting thresholds, eliminating the need for complicated threshold debugging after replacing monitoring scenes. The process of automatically determining the threshold screening results achieved a better effect than the artificially adjusted threshold. The method was applied to the deformation monitoring of the Mao county landslide in Sichuan and the PS point screening of the Feng mountain open-pit mine in Changping, and the feasibility of the algorithm was verified. The experimental results show that the three-threshold optimization algorithm can accurately determine the threshold and screen out the correct set of PS points.