3D Forest: An application for descriptions of three-dimensional forest structures using terrestrial LiDAR

Area of Science:

• Forestry science
• Geospatial analysis
• Remote sensing technology

Background:

• Terrestrial laser scanning (TLS) offers high-resolution 3D forest data.
• Extracting precise tree parameters from TLS data requires robust algorithms.
• Existing software solutions may lack comprehensive features or user-friendliness.

Purpose of the Study:

• To introduce 3D Forest, an open-source software for TLS data analysis.
• To provide an easy-to-use tool for extracting detailed forest and tree parameters.
• To present novel algorithms for automatic tree and crown segmentation.

Main Methods:

• Development of 3D Forest software with a graphical user interface.
• Implementation of algorithms for extracting stem positions, tree height, and diameter at breast height (DBH).
• Inclusion of advanced analyses for crown structure, inter-crown interactions, and automatic segmentation.

Main Results:

• 3D Forest accurately extracts tree height and DBH, with no significant difference from field measurements.
• The Randomized Hough Transform algorithm demonstrated robustness in DBH measurement against noise.
• The software quantifies complex crown parameters and spatial arrangements of trees.

Conclusions:

• 3D Forest is a valuable, open-source tool for detailed forest structure analysis using TLS data.
• The software provides accurate and reliable tree parameter extraction, aiding ecological research.
• Its user-friendly interface and advanced algorithms enhance the utility of TLS in forestry.