Research On The Detection Method Of Bridge Geometric Deviation Based On 3D Laser Scanning

As a new measurement method,3D laser scanning has the advantages of omnidirectional,high efficiency and high precision,which makes up for the shortcomings of the traditional single point measurement method,and gradually integrates into all aspects of the bridge construction monitoring field.Although the hardware of 3D laser scanner is developing rapidly,the corresponding laser point cloud data processing method is still in the exploration stage.In the aspect of bridge engineering,in the face of disordered and complex point cloud data,how to systematically complete the automatic processing of massive point cloud,quickly get the visualization deviation of bridge construction,and give full play to the advantages of 3D laser scanning technology,It is still an urgent problem to be solved.In this paper,combined with 3D laser scanning technology and BIM Technology,the BIM model is transformed into BIM theoretical point cloud model,the actual point cloud data collected by 3D laser scanning and BIM theoretical point cloud data are spatially matched,the relevant feature extraction algorithm is proposed,and the feature extraction of the actual point cloud model and BIM theoretical point cloud model is synchronous,so as to realize the rapid comparison of deviation and achieve the visualization of construction deviation At the same time,the multi utilization of BIM model is realized to avoid the problem of abstraction of deviation detection results.The main research contents are as follows(1)In order to explore the best scanning distance and azimuth range of high quality point cloud data,outdoor experiments were carried out to explore the influence of different scanning distance and scanning angle on the quality of point cloud.It was found that the scanning distance was more than 30 m and the scanning angle was greater than135 ° When scanning the object,the edge blur is found;In this paper,a registration method for the bridge holographic point cloud acquisition is proposed,which is not applicable to the general automatic point cloud registration method.The results show that the registration accuracy meets the requirements of the detection accuracy of bridge structure.(2)For the problem of low automation of feature extraction of component contour,a method of extracting component contour based on convex hull algorithm is proposed.Taking a light rail pier as an example,the synchronization deviation comparison is realized by combining BIM theory point cloud model,which verifies the feasibility of the algorithm.(3)In order to solve the problem of low automation and poor robustness of the point cloud data extraction axis and feature line of bridge structure,a new method of component axis automatic extraction based on Delaunay triangulation and a method of feature line automatic extraction based on virtual grid are proposed.The anti-difference estimation is introduced to the linear axis fitting in this paper,The robustness of the algorithm is improved by the method of weight selection iteration,and the comparison of the deviation of the alignment automation is completed by combining the actual bridge.(4)For the problems of the bridge members' defects are difficult to detect and detect,a defect automatic detection algorithm is proposed to find,locate,calculate the volume and calculate the area of the defects of the whole component automatically.Taking the indoor arch model as the research object,the accuracy of the defect detection algorithm is verified,and the feasibility verification of the algorithm is completed by combining with the actual bridge with local defects.