An App for Tree Trunk Diameter Estimation from Coarse Optical Depth Maps

Trunk diameter is a crucial indicator of a tree’s health and carbon sequestration capacity, making its measurement essential in both forest and urban environments. Traditional methods involve manual measurement with tools like measuring tapes or callipers, but recent advancements have introduced sophisticated technologies such as LiDAR and time-of-flight cameras, which offer detailed depth maps for precise measurements. However, these technologies are typically available only on specialized devices or high-end smartphones.

To address this limitation, this project has developed a mobile application that leverages coarse-grain depth maps generated by standard optical sensors, enabling it to run on most common Android devices. The app uses a state-of-the-art deep neural network to estimate trunk diameter from an RGB-D image (a combination of an optical image and its corresponding coarse depth map). The app offers a significant speed advantage, performing diameter measurements more than five times faster than traditional manual methods.