Abstract: Phase analysis based on high-coherence points can ensure the accuracy of ground-based SAR （Synthetic Aperture Radar） deformation measurement effectively. In the field of differential interferometry， the amplitude dispersion method is widely used， it can effectively select the PS （Permanent Scatterer） such as rocks and buildings as the high coherence points. However， for the vegetation slopes， the number of high-coherence points selected by this method is small， which is not conducive to differential interferometry. Stanford Method for Persisitent Scatterers （StaMPS） are commonly used in the field of spaceborne synthetic aperture radar （SAR） to solve the problem of high-coherence point selection of vegetation slope. In this paper， the applicability of the StaMPS method in the field of Ground-based synthetic aperture radar is explored， and a non-PS point is proposed to calculate the threshold of coherence coefficient， and a DS （Distributed Scatterer） selection technique is introduced to further improve the number of high coherence points. Experimental results on a vegetation slope show that， compared with the amplitude deviation method and the StaMPS method， the improved method not only increases the number of high coherent points， but also ensures the phase quality effectively.