On the 3D point clouds-palm and coconut trees data set extraction and their usages

Area of Science:

• Agricultural Science
• Computer Vision
• Geospatial Technology

Background:

• Drone imagery provides valuable data for field surveying and analytics.
• Current drone mapping software enables 3D object reconstruction.
• Open-source tools are increasingly prevalent in scientific research.

Purpose of the Study:

• To investigate the process of constructing 3D tree datasets using open-source software.
• To explore the applications of these 3D datasets, particularly in agriculture.
• To showcase the efficacy of open-source tools with case studies of palm and coconut trees.

Main Methods:

• Utilized OpenDroneMap for image processing and 3D reconstruction.
• Employed CloudCompare and Open3D for point cloud data manipulation and analysis.
• Developed 3D point cloud datasets for specific tree species.

Main Results:

• Successfully extracted 40 palm tree and 40 coconut tree point clouds.
• Demonstrated the practical application of the generated 3D datasets.
• Provided examples of usage in volume estimation and graph analytics.

Conclusions:

• Open-source software is effective for creating 3D tree datasets from drone imagery.
• The generated datasets have practical applications in agricultural analytics.
• This methodology offers a cost-effective and accessible approach to 3D geospatial data construction.