Abstract:The level of detail (LOD) modelling of vector and terrain data is individual, resulting in geometric and topological inconsistencies in simplification processes. The three dimension Douglas Peucker (3D_DP)algorithm can realize gradual discrete point selection through threshold, which is mainly used in DEM synthesis, and its simplified process is very suitable for the dynamic establishment of massive data sets. A new LOD modeling method based on 3D_DP algorithm is proposed to simplify the consistency of river network vector elements and DEM in this paper. The specific steps are as follows: Firstly, the "Bending Adjustment Index ( BAI )" is introduced to improve the 3D_DP algorithm, called the improved 3D_DP algorithm; Secondly, the DEM data is extracted into a 3D discrete point dataset, and the river line vector data is also converted into a discrete point dataset, assigned with elevation attributes, and merges with the DEM's 3D discrete points. The merged point datasets are equenced based on the importance of each point, which are computed by the improved 3D_DP algorithm. The order of deleting points is determined by the sequence and the corresponding model spatial errors are calculated with the deletion of points. Then, the DEM's 3D discrete points are constructed a Delaunay network with the river line as a mandatory constraint condition. The required triangulation is called in real time with the change of sight distance depending on the simple correspondence between screen projection error and model space error, and the unified LOD model for river line vector and DEM is established. The results show that the river's overall shape and the terrain's main features can be reserved under the same simplified factor based on the improved 3D_DP algorithm. The unified LOD model for the river network and DEM is feasible under the importance sequence of merged point datasets by the improved 3D_DP algorithm. Under the proper operation of data blocking, the rendering frame rate can meet practical application requirements.

Efficient View-Dependent LOD Control for Large 3D Unclosed Mesh Models of Environments

Efficient screen space approach for high complex dynamic scenes in real-time rendering

Simplification and View-Dependent LOD Control for Large 3D Mesh Models

Partition Based Real-Time Rendering Method for Large-Area Terrain Data

An Algorithm for LOD by Merging Near Coplanar Faces Based on Gauss Sphere.

Dynamic control of mesh LODs (levels of detail) by using a multiresolution mesh data structure

Adaptive voxels: interactive rendering of massive 3D models

Mesh-LOAM: Real-time Mesh-Based LiDAR Odometry and Mapping

DP-MVS: Detail Preserving Multi-View Surface Reconstruction of Large-Scale Scenes

Out-of-Core Constructing and Interactive Rendering of Multiresolution Representations for Massive Meshes

Level of Detail and Multi-Resolution Modeling for Virtual Prototyping.

Level of Detail and Multi-Resolution Modeling Techniques for Virtual Design and Prototyping

Dynamic Gaze-Controlled Levels of Detail of Polygonal Objects in 3-D Environment Modeling

Automatic Mesh and Shader Level of Detail

SMOOTH TRANSITION BETWEEN MODELS AT MULTIPLE LEVELS OF DETAIL

Distributed Refinement of Large-Scale 3D Mesh for Accurate Multi-View Reconstruction

A unified LOD model for river network and DEM based on an improved 3D_DP algorithm

View-dependent Continuous Level-of-detail Rendering of Terrain Model

High Quality Real-Time Rendering of Large Scale Point Model

DYNAMIC LEVEL OF DETAIL FRAMEWORK WITH TOPOLOGICAL STRUCTURE MODIFICATION

Feature-Based Visibility-Driven CLOD for Terrain