Slope Surface Displacement Monitoring Method And System Development Using Multi-view Registered Point Clouds

Slope engineering is one of the most common parts of hydropower engineering and highway engineering projects whose safety status is usually related to life and property.As the most apparent monitoring quantities to react the landslide and deformation trend,slope surface displacement is frequently measured by traditional devices such as GPS-RTK and total station.However,with the number of large-scale high steer slopes increasing,measuring and survey conditions are stricter,problems of traditional measuring method that costs more time,labor and sometimes exposes survey crew to danger occur as well.Hence,the non-contract measuring methods like unmanned aerial vehicle(UAV)and three-dimensional laser scanning are gradually appear.It has advantageous characteristics like fast,efficient and easy-operated which meets the requirements of slope surface monitoring.Meanwhile,the non-contract measuring methods are greatly limited when view is blocked,so the multi-view measuring is needed to ensure the accuracy.According to the above demonstration,this thesis proposes an improved slope surface monitoring method based on UAV photography to acquire the multi-view data and develops a slope surface monitoring system.Thesis researches mainly consist of following parts:(1)Correcting distortion and coordinate based on image control points.Acquiring UAV photography images,then comparing image control point size and characteristics in various flight altitude,correcting the point clouds position error due to distortion caused by camera lens,at the same time analyzing different algorithms in feature point recognition and matching to make preparation for point clouds registration.(2)Registering multi-view point clouds.Building the transformation relations among pixel coordinate system,image coordinate system,camera coordinate system and world coordinate system via image vision principle,using iterative closest point(ICP)to register point clouds from laser scanning and UAV photography,then using the unified point clouds to calculate surface displacement.(3)Calculating displacement and validating accuracy of proposed method.Based on the slope of real construction project,using UAV photography and laser scanning to generate point clouds,registering them and perform displacement calculation,comparing the result with traditional GPS-RTK measuring method on efficiency and accuracy to validate feasibility and accessibility of proposed method.(4)Developing a slope surface displacement monitoring system.Combining the above researches with software programming knowledge,framework and interface of the slope surface monitoring displacement system is designed and the system is developed which achieved surface reconstruction,displacement calculation and generation of contour,cloud picture and report.