Abstract: It has been realized that the points collected twice by the laser scanner for the same target do not coincide, and the landslide displacement cannot be quickly determined by direct comparison of the point cloud. Because the position of a single point cloud is uncertain and the density of regional point clouds is stable, the density of the point cloud can be used as the characterization of the deformation of the landslide surface. The article proposes a landslide displacement monitoring method based on point cloud density characteristics. First, the discrete three-dimensional point cloud is converted into a two-dimensional density image. Then, the particle image velocimetry technique is used to analyze the correlation between the two point cloud density images before and after displacement, thereby to calculate the relative displacement value of each subset in the raster image. When the entire displacements of each subset are calculated, the plane displacement field of the target area is obtained. The indoor block movement test shows that the calculation accuracy of this method is affected by the deformation gradient, which will produce a certain degree of error at the place where the ground surface changes drastically, and the subset correlation coefficient cannot reach 1.0. In the displacement monitoring of the Huangzangsi landslide, the method was used to identify the changing area of the slope, and the plane displacement field of the landslide was calculated. The calculation results intuitively reflect the deformation of the landslide surface, which verifies the practicability of the proposed method.

Key words: 3D laser scanning technology, particle image velocity measurement, landslide monitoring, point cloud comparison