Three-Dimensional Point Cloud Feature Change Detection Algorithm Of Open-pit Mine Based On Discrete Curvature Analysis

Open-pit mine is an industrial production site produced by human intervention,easy to produce human and nonhuman changes.Landslide and stairs are the main model structure of open-pit characteristics,the changes of landslide and stairs can clearly describe by using the changes of stairs feature lines,but changes of stairs feature lines correspond to the changes of landslide and stairs,and the sliding and change have affected safety production of open-pit mine.Traditional landslide monitoring means of open-pit mine are mainly that fixed finite monitoring points on the side slope are monitored by total station measuring robot,to obtain the results of change detection,but detection density is very low,small area and low efficiency.This paper analyzes the status quo of the research on the feature extraction of point cloud model,stairs feature line extraction in open-pit mine are still human computer interactive method at present.Research on the automatic extraction of stairs feature lines in open-pit mine 3D point cloud model is studied.The main contents of the study includeThis paper have solved two key algorithms study of discrete Curvature analysis based on point cloud triangular mesh model and feature line extraction on the open-pit mine.First of all,the original point cloud data were resampled,coordinate transformation of the model is proposed,point cloud filtering is proposed based on iterative linear least squares interpolation filtering algorithm,to filter out vegetation and construction information of point cloud data.Through establishing Delaunay triangular mesh model of the point cloud data,to obtain vertex's curvature analysis.Vertex 's normal vector,Gaussian curvature,principal curvature and mean curvature parameters is calculated,to get the visual curvature value map of open-pit mine model.Finally,through the complex method of Morse-Smale(MS),setting index function and significant degree of feature line,the feature line are acquired and smoothed.Research on multi-phase point cloud data of open-pit mine,bench feature line of open-pit mine,value of the slope,value of the aspect,value of contour are acquired using these parameters to detect change.Such as,the characteristics of the bench,the change of the slope,the aspect,the height difference and the volume detected.This paper is studying 4-dimensinal change detection of open-pit mine through multiple change detection parameters.Based on the algorithm research,in this paper west open-pit mine 2013-2014 two laser scanning data and qidashan open-pit mine multiphase point cloud data are used for experiments.The correctness of the algorithm in this dissertation were conducted.The result of the study prove that there are certain application value in landslide change detection and open-pit mine safety production field,based on the discrete curvature analysis of open-pit mine 3D point cloud feature change detection.