基于无人机遥感影像的滑坡形态变形特征： 以陇南白龙江流域泻流坡滑坡为例

2020年8—9月陇南白龙江流域泻流坡滑坡持续发生变形，严重威胁坡脚居民生命财产安全。以泻流坡滑坡为研究对象，基于高分辨率无人机影像提取了泻流坡滑坡滑动前后的正射影像和数字地表模型，利用COSI-Corr软件对滑动前后的两期影像进行相关性分析，计算得到滑坡表面特征点的偏移量信息，并通过DSM(digital surface model)的方法获取垂直向的变化量，最后对泻流坡滑坡的运动和物质变化进行了解译和分析，探讨了中高山峡谷区堆积层滑坡形成演化过程及破坏机制。研究结果表明：泻流坡滑坡变形区主要为滑源区左侧中下部和右侧次级滑坡，垂向最下滑移量5.89 m,平均水平位移6.24 m,滑源区近42%区域发生了形变；2020年6—9月，泻流坡滑坡滑动体积33 871 m³,堆积体积10 215 m³,降雨侵蚀体积23 656 m³;破坏模式为倾倒拉裂-蠕滑复合型滑坡。无人机高精度数据不仅可以清楚直观地识别滑坡形变破坏迹象，还可以进行地表垂直位移、水平位移、体积变化及滑动前后剖面的计算，为抢险救灾、科学决策提供了重要依据，具有广...

Study on the morphological deformation characteristics of landslide based on UAV imagery：taking Xieluopo landslide in Bailong river basin of Longnan as an example

From August to September 2020, the Xieliupo landslide in the Bailong River basin in Longnan continued to deform, seriously threatening the life and property safety of residents at the foot of the slope. In this paper, Xieliupo landslide is taken as the research object. Based on the high-resolution UAV imagery, the orthophoto images and digital surface models before and after the sliding of the Xieliupo landslide are extracted, of which the correlation is analyzed by COSI-Corr software. The offset information of the characteristic points on the landslide surface is calculated, and the vertical variation is obtained through the DSM difference method. Finally, the movement and material change of the Xieliupo landslide are understood and analyzed, and the formation and evolution process and failure mechanism of the colluvial landslide in the middle and high mountains and valleys are discussed. The research results show that the deformation area of Xieliupo landslide is mainly the left middle and lower part of the slip source area and the right secondary landslide. The maximum vertical sliding displacement is 5.89 m, the average horizontal displacement is 6.24 m, and nearly 42% of the slip source area has been deformed. From June to September 2020, the sliding volume of torrential slope landslide is 33871 m3^{, the accumulation volume is 10215} m3^, the rainfall erosion volume is 23656 m3^{, and the} failure mode is a creeping – tension type thrust load-caused landslide. UAV high-precision data can not only clearly and intuitively identify the signs of landslide deformation and damage, but also calculate the surface vertical displacement, horizontal displacement, volume change and the profile before and after sliding, which provides an important basis for rescue and disaster relief, scientific decision-making, and has broad application prospects.