Landslide Identification and Gradation Method Based on Statistical Analysis and Spatial Cluster Analysis

As a type of earth observation technology, interferometric synthetic aperture radar (InSAR) is increasingly widely used in the field geological disaster detection. However, application InSAR low-coherence areas, such as alpine canyon areas and vegetation coverage subject to considerable limitations. How accurately identify landslides from measurement data these remains several challenges shortcomings. Based on statistical analysis spatial cluster analysis, this paper, we propose an automatic landslide identification gradation method suitable for areas. The proposed combines small baseline subset (SBAS-InSAR) interferogram stacking (stacking-InSAR) obtain deformation map study area, using extract regions polygons screening model (LSM) based multivariate features screen reduce interference noise identification, addition proposing (LGM) signum function grade identified provide support distinguish with different degrees. was applied upper section Jinsha River basin, 47 potential were identified, including 15 high-risk 13 endangering villages. experimental results show that can hierarchically providing hazard investigation agencies local departments.