- Titel
- A new open source software to design models for automatic 3D point cloud classification in environmental studies
- Untertitel
- cLASpy_T
- Zitierfähige Url:
- https://nbn-resolving.org/urn:nbn:de:bsz:14-qucosa2-984920
- Konferenz
- 5th Virtual Geoscience Conference. Dresden, 21. - 22. September 2023
- Quellenangabe
- VGC 2023 - Unveiling the dynamic Earth with digital methods
Herausgeber: Helmholtz-Institut Freiberg für Ressourcentechnologie
Herausgeber: Technische Universität Dresden
Erscheinungsort: Dresden
Erscheinungsjahr: 2023
Seiten: 20-21 - Erstveröffentlichung
- 2023
- DOI
- https://doi.org/10.25368/2025.162
- Abstract (EN)
- We introduce a new open source software, cLASpy_T, that helps classify environmental 3D point clouds. cLASpy_T means ‘Tools for classification of LAS files with Python’. Its primary mission is to assist those who have to try out various machine learning algorithms on their point clouds. These could be derived from photogrammetry or LiDAR scanners enhanced with geometric features and/or spectral information from colors (RGB), multi- or hyperspectral bands. cLASpy_T transforms LAS files with additional features to make it compatible with the scikit-learn library. The software is divided into three modules: ‘training’, ‘predict’ and ‘segment’. These modules respectively train supervised classification models, make predictions with these models, and segment point clouds into clusters.
- Freie Schlagwörter (DE)
- Maschinelles Lernen, Klassifizierung, Umweltdaten, Photogrammetrie, luftgestütztes LiDAR, hyperspektrale Daten
- Freie Schlagwörter (EN)
- machine learning, classification, environmental data, photogrammetry, airborne LiDAR, hyperspectral data
- Klassifikation (DDC)
- 550
- Klassifikation (RVK)
- RB 10081
- Herausgeber (Institution)
- Helmholtz-Institut Freiberg für Ressourcentechnologie, Freiberg
- Technische Universität Dresden, Dresden
- Version / Begutachtungsstatus
- publizierte Version / Verlagsversion
- URN Qucosa
- urn:nbn:de:bsz:14-qucosa2-984920
- Veröffentlichungsdatum Qucosa
- 15.08.2025
- Dokumenttyp
- Konferenzbeitrag
- Sprache des Dokumentes
- Englisch
- Lizenz / Rechtehinweis
CC BY 4.0