Goebel, Mona ; Schmidt, Jakob ; Iwaszczuk, Dorota (2023)
Segmentation of Forest Vegetation Layers Based on Geometric Features Extracted from 3D Point Clouds.
In: Allgemeine Vermessungs-Nachrichten (AVN), 130 (5)
Artikel, Bibliographie
Kurzbeschreibung (Abstract)
The analysis of forest vegetation at lower heights, up to 2 m, is the focus of this work, while previous approaches primarily focused on trees and their stems. We calculated geometric metrics of point clouds, based on airborne, unmanned, and mobile laser scanning, to segment different vegetation growths and densities. Our results show that metrics based on eigenvalues such as planarity, linearity, and sphericity, as well as normal change rate are useful to differentiate forest layers in our scenario. Volume density is ineffective here as it is highly dependent on the data collection method. Roughness and principal component analysis in main direction (PCA1) do not show significant differences in vegetation growth and height. This research lays the foundation for the usage of geometric metrics to highlight growth and density changes in 3D vegetation and to differentiate forests into different layers without extensive processing.
| Typ des Eintrags: | Artikel | ||||
|---|---|---|---|---|---|
| Erschienen: | 2023 | ||||
| Autor(en): | Goebel, Mona ; Schmidt, Jakob ; Iwaszczuk, Dorota | ||||
| Art des Eintrags: | Bibliographie | ||||
| Titel: | Segmentation of Forest Vegetation Layers Based on Geometric Features Extracted from 3D Point Clouds | ||||
| Sprache: | Englisch | ||||
| Publikationsjahr: | 2 Oktober 2023 | ||||
| Ort: | Berlin | ||||
| Verlag: | Wichmann | ||||
| Titel der Zeitschrift, Zeitung oder Schriftenreihe: | Allgemeine Vermessungs-Nachrichten (AVN) | ||||
| Jahrgang/Volume einer Zeitschrift: | 130 | ||||
| (Heft-)Nummer: | 5 | ||||
| URL / URN: | https://gispoint.de/artikelarchiv/avn/2023/avn-ausgabe-05202... | ||||
| Zugehörige Links: | |||||
| Kurzbeschreibung (Abstract): | The analysis of forest vegetation at lower heights, up to 2 m, is the focus of this work, while previous approaches primarily focused on trees and their stems. We calculated geometric metrics of point clouds, based on airborne, unmanned, and mobile laser scanning, to segment different vegetation growths and densities. Our results show that metrics based on eigenvalues such as planarity, linearity, and sphericity, as well as normal change rate are useful to differentiate forest layers in our scenario. Volume density is ineffective here as it is highly dependent on the data collection method. Roughness and principal component analysis in main direction (PCA1) do not show significant differences in vegetation growth and height. This research lays the foundation for the usage of geometric metrics to highlight growth and density changes in 3D vegetation and to differentiate forests into different layers without extensive processing. |
||||
| Alternatives oder übersetztes Abstract: |
|
||||
| Freie Schlagworte: | Metric, mobile laser scanning, airborne laser scanning, UAV laser scanning, forest, vegetation, segmentation | ||||
| Fachbereich(e)/-gebiet(e): | 13 Fachbereich Bau- und Umweltingenieurwissenschaften 13 Fachbereich Bau- und Umweltingenieurwissenschaften > Institut für Geodäsie 13 Fachbereich Bau- und Umweltingenieurwissenschaften > Institut für Geodäsie > Fernerkundung und Bildanalyse |
||||
| Hinterlegungsdatum: | 10 Okt 2023 05:07 | ||||
| Letzte Änderung: | 23 Okt 2023 11:28 | ||||
| PPN: | 512643598 | ||||
| Export: | |||||
| Suche nach Titel in: | TUfind oder in Google | ||||
 |
Frage zum Eintrag |
Optionen (nur für Redakteure)
 |
Redaktionelle Details anzeigen |
Drucken
Impressum