Tuorial

The DeepTrees python library allows to train and implement tree crown delineation dee learning models using PyTorch. You are also able to analyze the predictions to derive tree metrics using this library, provide an end-to-end anylsis for downstream tasks. The libary incorporates the mode architecture from Freudenberg et al., 2022, which is based on the U-Net architecture (Figure 1).

The following mindmap provides an overview of what the DeepTrees library can do and how it can be used in your tree analysis workflows:

Installation

Configuration

Predictions and Analysis

Pre-trained models for tree segmentation

Predict on tiles

Tree masks

Distance transform maps

Pixel-wise entropy maps

Analyzing tree metrics

Training

Fine-tuning pre-trained models

Active learning

Training your own model

References

Tree Segmentation

UNet Segmentation

Freudenberg, M., Magdon, P. and Nölke, N., 2022. Individual tree crown delineation in high-resolution remote sensing images based on U-Net. Neural Computing and Applications, 34(24), pp.22197-22207.

Website Code