基于凹包与最小外接矩形的地涵点云边界提取与拟合OA

During the long-term operation of the ground culvert,the problem of sedimentation in the ground culvert seriously affects the safety and operation of the hydraulic facilities.Traditional pipeline detection methods have limitations in detecting complex ground culverts.Although sonar technology can produce high-precision point cloud data,extracting complex point cloud boundaries remains a difficult problem.Given the shortcomings of existing boundary-extraction methods for complex ground-culvert boundaries,this paper proposes a technique that combines the concave hull and the minimum enclosing rectangle.This method first converts cross-sectional sonar data into point clouds,removes noise,then uses the concave hull algorithm to extract boundaries that retain concave features,and finally uses the minimum enclosing rectangle to fit and optimize.By testing the sonar data from the overpass ground culvert,this method can effectively restore boundary information,optimize global consistency,and is applicable to point clouds of different densities,providing technical support for detecting ground culvert sedimentation and related applications.