AUTOMATIC TREE DETECTION/LOCALIZATION IN URBAN FOREST USING TERRESTRIAL LIDAR DATA

Author(s)

dos Santos, Renato Cesar ; da Silva, Matheus Ferreira ; Tommaselli, Antonio Maria G. ; Galo, Mauricio

• Document type: Journal article
• Source: IGARSS 2024-2024 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM, IGARSS 2024; v. N/A, p. 4-pg., 2024-01-01.

Abstract

Individual tree detection is essential task to access relevant parameters at the tree scale, such as: diameter at breast height (DBH), first branch height, and tree height. In this context, we propose an automatic tree detection/localization approach based on trunk geometry, i.e., on vertical continuity, not requiring preprocessing stages (ground filtering, point cloud normalization, classification) or training samples, as in some classes of methods. The performance of the proposed approach was evaluated using LiDAR data acquired by a terrestrial laser scanning (TLS) system in an urban forest. Obtained results indicated the potential of the proposed approach, resulting in an Fscore of 98% and a RMSEXY of 15 cm. (AU)

• FAPESP's process: 24/04106-2 - 2024 IEEE International Geoscience and Remote Sensing Symposium - IGARSS
• Grantee: Renato César dos Santos
• Support Opportunities: Research Grants - Meeting - Abroad
• FAPESP's process: 21/06029-7 - High resolution remote sensing for digital agriculture
• Grantee: Antonio Maria Garcia Tommaselli
• Support Opportunities: Research Projects - Thematic Grants
• FAPESP's process: 22/11647-4 - Multi-user equipment approved in grant 2021/06029-7: Terrestrial Laser Scanner FARO Focus Premium 70
• Grantee: Antonio Maria Garcia Tommaselli
• Support Opportunities: Multi-user Equipment Program