Szczegóły publikacji
Opis bibliograficzny
New approach towards Digital Elevation Model data generalisation using the Douglas-Peucker algorithm and Delaunay triangulation based on characteristic boundary points / Mariusz Zygmunt, Marta RÓG // Measurement ; ISSN 0263-2241 . — 2026 — vol. 260 art. no. 119849, s. 1–12. — Bibliogr. s. 11–12, Abstr. — Publikacja dostępna online od: 2025-11-25
Autorzy (2)
- Zygmunt Mariusz
- AGHRóg Marta
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 164936 |
|---|---|
| Data dodania do BaDAP | 2026-02-11 |
| Tekst źródłowy | URL |
| DOI | 10.1016/j.measurement.2025.119849 |
| Rok publikacji | 2026 |
| Typ publikacji | artykuł w czasopiśmie |
| Otwarty dostęp | |
| Czasopismo/seria | Measurement |
Abstract
This article presents a novel approach to the extraction of characteristic points for initial triangulation in a multi-pass refinement method. It is tailored particularly for the generalisation of Digital Elevation Model (DEM) data from large sets of points, e.g. 3D point cloud-based GRID data sets obtained via laser scanning. The entire workflow involves two main stages: extraction of characteristic points along dataset boundaries using the Douglas-Peucker algorithm to prepare for initial triangulation, and (2) recursive Delaunay triangulations performed to satisfy a specified Z-tolerance across the entire study area. Points are added using either a sequential or parallel processing approach. The algorithm was tested in three study areas, with three Z-tolerance values. The workflow was compared with tools available in ESRI software, LAStools, TerraScan, and SAGA GIS. Results demonstrate the method's effectiveness in TIN data generalisation, with significant data reduction (up to 56% using a sequential variant, compared to the ESRI algorithm), and a reduced number of iterations needed to obtain the final model. RMSE values were within acceptable limits and followed a normal distribution in all but one case, which is explained further in the article. Unlike many existing approaches, our algorithm allows seamless merging of adjacent TINs as it introduces the step of characteristic boundary points extraction. The algorithm is transparent, reproducible and straightforward to integrate into GIS workflows. © 2025 Elsevier Ltd.