ReforesTree: A Dataset for Estimating Tropical Forest Carbon Stock with Deep Learning and Aerial Imagery

TL;DR

It is shown that a deep learning-based end-to-end model using individual tree detection from low cost RGB-only drone imagery is accurately estimating forest carbon stock within official carbon offsetting certification standards, and the baseline CNN model outperforms state-of-the-art satellite-based forest biomass and carbon stock estimates for this type of small-scale, tropical agro-forestry sites.

Abstract

Forest biomass is a key influence for future climate, and the world urgently needs highly scalable financing schemes, such as carbon offsetting certifications, to protect and restore forests. Current manual forest carbon stock inventory methods of measuring single trees by hand are time, labour, and cost-intensive and have been shown to be subjective. They can lead to substantial overestimation of the carbon stock and ultimately distrust in forest financing. The potential for impact and scale of leveraging advancements in machine learning and remote sensing technologies is promising but needs to be of high quality in order to replace the current forest stock protocols for certifications. In this paper, we present ReforesTree, a benchmark dataset of forest carbon stock in six agro-forestry carbon offsetting sites in Ecuador. Furthermore, we show that a deep learning-based end-to-end model using individual tree detection from low cost RGB-only drone imagery is accurately estimating forest carbon stock within official carbon offsetting certification standards. Additionally, our baseline CNN model outperforms state-of-the-art satellite-based forest biomass and carbon stock estimates for this type of small-scale, tropical agro-forestry sites. We present this dataset to encourage machine learning research in this area to increase accountability and transparency of monitoring, verification and reporting (MVR) in carbon offsetting projects, as well as scaling global reforestation financing through accurate remote sensing.

Figures (7)

• Figure 1: Drone imagery of each site of the ReforesTree dataset with a resolution of 2cm/px. The red dots are the locations of the trees measured in field surveys, plotted to make clear that the coverage of drone images were larger than the field measured area.
• Figure 2: The standard procedure for calculating the correct amount of carbon offsets to be certified for a reforestation project. The tree metrics are collected from manual forest inventory.
• Figure 3: The raw data and data processing pipeline for the ReforesTree dataset, resulting in labels matched to bounding boxes per tree.
• Figure 4: Bounding box annotations per tree, as a result of fine-tuned DeepForest tree crown detection and manual cleaning. Red boxes represent banana trees and blue boxes represent other species.
• Figure 5: This figure represents the count of species family groups for each of the sites. All sites have trees of all species family groups, but cacao and banana are over represented.