Estimating the attributes of urban trees using terrestrial photogrammetry

Area of Science:

• Forestry and Urban Ecology
• Geospatial Analysis
• Remote Sensing

Background:

• Accurate measurement of urban tree attributes is crucial for management and ecological studies.
• Traditional methods can be time-consuming, costly, or destructive.
• Terrestrial photogrammetry (TP) offers a potential non-destructive alternative for 3D data acquisition.

Purpose of the Study:

• To evaluate the accuracy of TP for estimating 2D and 3D attributes of urban trees at seedling, single tree, and plot levels.
• To compare TP-derived measurements with ground-truth data.
• To assess the feasibility of TP as an alternative to laser scanning.

Main Methods:

• Utilized Structure-from-Motion with Multi-View Stereo-photogrammetry (SfM-MVS) to generate point clouds and 3D models from TP data.
• Collected data across three levels: seedlings, single trees, and sample plots.
• Validated estimated tree attributes against measured values.

Main Results:

• Root Mean Square errors (RMSE%) for diameter at breast height (DBH) and mid-trunk were below 2% across all levels.
• RMSE% for total height and crown height did not exceed 4% and 5%, respectively.
• Stem volume estimation showed lower errors for trees in sample plots (14.37%) compared to seedlings (20.99%).

Conclusions:

• TP provides a highly accurate and cost-effective method for assessing urban tree dimensions.
• The technique is suitable for various tree stages, from seedlings to mature trees in plots.
• TP presents a practical, non-destructive alternative to laser scanning for urban forest inventory.