Research On 3D Modeling Of High Cliffs And Extraction Of Rock Mass Joint Information

The extraction of joint information in the engineering geological survey of high steep cliffs is an important content and the basis for the analysis and evaluation of engineering geological problems.The high and steep cliffs have strong mountains,complex terrain,and uneven rock masses,which are difficult for geological personnel to reach.It is difficult to collect joint information of outcrop rocks using traditional geological survey methods such as compasses and rulers.In addition,traditional geological survey methods have a large workload and low efficiency in large-scale high-cliff geological surveys over hundreds of meters.Three-dimensional laser scanning technology can digitally collect three-dimensional information of dangerous,difficult,difficult and delicate areas.It has certain advantages in data collection of complex scenes.It is a non-contact active measurement technology.It has high efficiency and accuracy.It can be used in computers.Display and analyze the collected 3D point cloud.Therefore,3D laser scanning technology can be used for high steep cliff data collection,but the 3D laser scanner will be affected by the field of view angle limitation and on-site environmental factors.Multi-station scanning cannot obtain complete high cliff surface information.In order to obtain complete three-dimensional data of high cliffs,it is supplemented by the drone tilt photogrammetry technology,which can collect data from top to bottom,wide field of view,and bottom-up collection method with three-dimensional laser scanning technology Combined,you can get comprehensive high cliff data.Then carry out fusion research on the collected multi-source data to obtain complete three-dimensional data on the surface of high cliffs.Finally,a three-dimensional model of high cliffs is established,and the required rock mass joint information is extracted from the fusion data and the three-dimensional model.The main research contents and results of this article are as follows:(1)"Point cloud + point cloud" fusion research.First,use 3D laser scanning technology and UAV tilt photogrammetry technology to collect high cliff data;Then,preprocess the laser point cloud to obtain a point cloudconsistent with the spatial information of the high cliff;Secondly,process the image data of the drone tilt photogrammetry to obtain the image point cloud;Finally,the image point cloud is fused into the laser point cloud to obtain the complete multi-source point cloud data of high steep cliffs,and the three-dimensional modeling of the fused multi-source point cloud data is studied.The fusion method uses a combination of unified coordinate system method,local feature descriptor and improved ICP algorithm.(2)"Model + model" fusion research.Three-dimensional reconstruction of the image point cloud and laser point cloud are carried out separately,and then the two models are fused together to form a "model + model" fusion study.The fusion method of the model space position uses a unified coordinate system combined with an improved ICP algorithm,and finally fuses the grid to make it a whole three-dimensional model.(3)Perform precision analysis on different modeling methods and compare the advantages and disadvantages of modeling.The experimental results show that the absolute mean error of the final modeling fusion obtained by using the "point cloud + point cloud" fusion method and the "model + model" fusion method is within 0.067 m.The multi-source point cloud fusion modeling has high precision,slow modeling speed,and high computer resources;the "model + model" fusion modeling method is fast and the computer resources are low,but the absolute accuracy is relatively higher than that of multi-source point cloud fusion modeling.Slightly lower.(4)Research on extraction of joint information of high steep cliffs.In the fused multi-source point cloud,the fuzzy C-means clustering method is used to extract the joint information of outcrop rock mass,and the regional growth algorithm is used in the 3D model to extract the joint information of outcrop rock mass.The experimental results show that these two methods can effectively extract the occurrence and joint information of rock mass.It can solve the problem of data collection and automatic extraction of joint information in geological survey.This method can be applied to large-area regional geological adjustment to improve production efficiency.