基于多时相光学遥感影像亚像素相关的边坡监测——以溪洛渡电站为例

Sub-pixel correlation-based slope monitoring using multi-temporal optical remote sensing images:a case study of Xiluodu hydropower station

水电站高边坡监测是水电站灾害防治中的关键问题,基于卫星遥感技术进行库区滑坡监测是解决该问题的重要手段之一。利用卫星影像亚像素相关性算法获取地表形变位移场,能够克服SAR影像失相关因素,在水电站边坡监测中具有重大的应用潜力。本文以溪洛渡电站2015-2019年5期Google Earth影像和2019-2022年4期Sentinel-2影像为数据源,采用相位相关算法计算了边坡形变量,通过构建一次多项式曲面拟合模型去除轨道误差等趋势性误差,获取了2016、2017和2019年溪洛渡电站下游边坡形变值。分析显示,两种影像提取得到的边坡形变量具有相同的变化趋势,均与谷幅实测数据吻合较好。本文结果验证了基于多时相遥感数据将亚像素相关性匹配技术运用于大型水电站边坡形变监测的可行性。

The monitoring of high slope in hydropower plant is a key issue in disaster prevention and control of hydropower plant.The use of satellite remote sensing technology for reservoir landslide monitoring is an important way to solve the problem.The use of subpixel correlation algorithm to obtain the ground deformation displacement field can overcome the phase decorrelation factors in SAR images,and has great potential application in the monitoring of hydropower station slopes.In this paper,Google Earth images from 2015 to 2019 and Sentinel-2 images from 2019 to 2022 were used as data sources,and the slope deformation variables were calculated by the phase correlation algorithm.The trend error such as orbit error was removed by constructing a first-order polynomial surface fitting model,and the slope deformation values of the downstream slope of Xiluodu Station in 2016,2017,and 2019 were obtained.The analysis showed that the slope deformation values obtained from two images had the same trend,and were in good agreement with the measured data of valley width.The results of this paper verified the feasibility of applying the subpixel correlation matching technique,based on multi-temporal remote sensing data,for monitoring the slope deformation of large hydropower plants.