An intertidal zone salt marsh vegetation point cloud filtering method based on an XGboost algorithm
By combining UAV multi-line array LiDAR technology with the XGboost algorithm, and utilizing point cloud intensity, incident angle, distance, and normal vector features, a point cloud filtering model for intertidal salt marsh vegetation is constructed. This solves the problem of insufficient accuracy in obtaining terrain information in salt marsh vegetation-covered areas in traditional methods, and achieves high-precision point cloud filtering and terrain information acquisition.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGZHOU URBAN PLANNING & DESIGN SURVEY RES INST
- Filing Date
- 2023-10-09
- Publication Date
- 2026-07-24
AI Technical Summary
Traditional topographic surveying methods struggle to obtain high-precision digital elevation models in intertidal regions, especially in areas with dense salt marsh vegetation, making it difficult to accurately separate ground points from non-ground points.
Three-dimensional point cloud data of the intertidal zone was acquired using UAV multi-line array LiDAR technology. Point cloud intensity, incident angle, distance, elevation and normal vector were used as input features. The XGboost algorithm was used to construct a classification model of ground points and non-ground points. The objective function was optimized through gradient boosting strategy to achieve high-precision filtering of point cloud data.
It achieves high-precision filtering of point clouds of intertidal salt marsh vegetation, which can accurately separate ground points and non-ground points, improves the accuracy and applicability of terrain information acquisition, and avoids a complex intensity correction process.
Smart Images

Figure CN117370800B_ABST