| The fold,fracture and breakage of solid rock under stress will lead to the instability of rock mass structure,landslide or collapse,which will seriously endanger the safety of human production and life.External space form of the rock mass are the external manifestations of the internal force,including the structural plane of rock mass,the rock mass discontinuity trace of the state and the spatial distribution of related parameters,is to analyze the important reference index of the stability of rock mass,how to quickly and accurately obtain and analysis spatial attribute information of rock mass and has become a hot research topic in the field of rock mass slope problem.The traditional contact measurement method in the actual operation,due to the measurement personnel’s own level and the complexity of the site situation,will produce a large error to the collection results,and time-consuming,laborious,dangerous.Non-contact measurement technology,such as unmanned aerial vehicle(UAV)close shot photogrammetry,3D laser scanner,digital camera close range measurement,etc.,can be in contact with the surface of the rock mass cases,rapid access to rock mass characterization data,more,low risk of the save labour when the province,compared with the traditional method is suitable for high and steep slope,waters along the slope and modern large-scale geotechnical structure face safe and efficient access to the original information.Based on the 3D point cloud data with texture obtained by non-contact measurement technology,this paper researches on how to refine and classify the structural plane,how to extract discontinuous trace and calculate relevant parameters by combining the color information and geometric information of the 3D point cloud.The main research contents are as follows:1)Automatic extraction and parameter calculation of refined classification of rock mass discontinuities.Based on the preliminary group extraction of discontinuities according to their occurrence,a classification method of refining discontinuities of rock mass based on density clustering was proposed to achieve automatic classification of discontinuities of the same group.Through the flow and algorithm of projection transformation,scattered point fitting and straight line normalization,the spacing of structural planes and the volume joint number of rock mass can be obtained.2)Automatic extraction and parameter calculation of discontinuity trace of rock mass.A point cloud based on color information and geometry information of trace automatic extraction method,can be carried out on the rock mass discontinuity trace geometry and texture of double extraction,gain potential discontinuity trace points,and the normalization of local linear smoothing technology to refine the potential continuity trace point,get more accurate discontinuity trace point;A bidirectional connection algorithm based on local vector buffer and connectivity judgment was proposed to optimize the connection effect of discontinuity traces on rock surface.The calculation and analysis methods of mass density and average track length of rock mass are designed.3)A prototype system for rock mass surface point cloud processing and analysis is designed and developed,which realizes the functions of structural plane refinement and classification,automatic extraction of discontinuous trace,spatial information calculation and statistical analysis of structural plane and trace,etc.Through four groups of real case data verify the validity of the method in this paper,the method of extracting results and contrastive analysis of the existing literature method results show that the proposed classification of rock mass structural plane refining extract and rock mass discontinuity trace extraction method is rapid,effective and high degree of automation,can be used as a traditional artificial contact measurement method for rock mass surface information of beneficial complement,It also provides a new idea for the extraction of discontinuity trace and parameter calculation of rock mass discontinuities and the research and development of application software. |
