LiDAR Extraction and Accuracy Analysis of Cross-Sections from Constrained by Quasi-Geoid Model

To build a high-fidelity digital twin of water network areas, it is necessary to accurately obtain and analyze the "slice" data of fixed sections of key parts such as embankments and river channels. However, the traditional photogrammetric triangulation height measurement method is restricted in the Sankou flood channel area due to frequent settlement of embankment markers, poor reliability of control points' benchmarks, weak continuity of historical height benchmarks, and the obstruction of side beaches and significant errors in long-distance multi-station connection measurements. Therefore, it is urgent to propose an efficient and accurate method for obtaining fixed section data suitable for this area. In response to the above problems, this paper proposes a new method for intelligent analysis of fixed sections based on the quasi-geoid constraint of airborne point clouds. Firstly, a robust estimation least squares algorithm is used to construct a refined quasi-geoid model of the Sankou flood channel area, establishing a unified and stable height benchmark throughout the region. Then, with this refined model as a constraint, the airborne point clouds are precisely processed to achieve intelligent extraction and generation of fixed section "slices". The internal and external consistency accuracy verification and the comparison with historical benchmarks show that this method effectively ensures the continuity of the height benchmark and the accuracy of the measurement results, while significantly improving the efficiency and reliability of obtaining fixed section data. Therefore, this method provides high-quality basic data support for the analysis of sedimentation and erosion in the Sankou flood channel, effectively solving the technical problems faced by traditional measurement methods in this area, and has important practical significance for the construction of digital twins in water network areas.