Experimental Research On Displacement Monitoring Of Simulated Dump By Close Range Photogrammetry Of Sequence Images

Large-scale projects such as open-pit mining,road construction,water conservancy and hydro-power will form a large number of artificial slopes,and the slopes will slip under natural or human factors,which may cause serious harm to human production safety and life safety.Landslide monitoring may be possible to combat the creeping,sliding landslide deformation and provide technical support,as displacement,deformation prediction and forecasting trends slopes,provide the basis for further development of disaster reduction programs.A sequence image landslide monitoring scheme is proposed in this article that can be used for real-time,long-distance transmission.This scheme mainly uses the method of matching the adjacent images before and after the sliding,and uses the image-side displacement field of the same name to calculate the object-side displacement vector field.The cracks generated by the displacement of objects can provide a basis for the detection of landslide precursors,which means that the time,direction and distance of landslides can be predicted.At the same time,the experimental study of the placement scheme of the control points for the displacement field of the object has guiding significance for the establishment of the actual landslide monitoring and control field.Therefore,two aspects of the study are as follows in this paper:1.A landslide crack monitoring method was proposed.Landslide crack detection scheme proposed couples two-dimensional field displacement vector model based on image feature points and neighboring image pixel value grayscale jump test model,which can define landslide cracks and landslide areas at the same time.The experimental results show that the landslide areas and landslide cracks can be delineated according to this scheme,providing a reliable basis for landslide monitoring and early warning and formulating disaster reduction plans.2.A slope displacement monitoring program was designed.This scheme firstly selects the image matching algorithm according to the characteristics of landslide monitoring,and then obtains the same name point of the neighboring image in the monitoring area,and calculates the displacement vector of the same-name point image.Finally,according to the accuracy evaluation result,the best analytical algorithm is chosen to obtain the object displacement vector field.At the same time,the law of the accuracy of the solution increases with the number of control points is studied.The final research results show that the SIFT image matching algorithm is the best matching algorithm for slope displacement monitoring,and the direct linear transformation method has the highest solving accuracy when converting the image coordinates into the ground coordinates,and the stable calculation accuracy is obtained when the number of control points reaches 9,which provides experimental basis for the actual slope monitoring program design.