Deep Transformer-Based Network Deforestation Detection in the Brazilian Amazon Using Sentinel-2 Imagery

Abstract

Deforestation poses a critical environmental challenge with far-reaching impacts on climate change, biodiversity, and local communities. As such, detecting and monitoring deforestation are crucial, and recent advancements in deep learning (DL) and remote sensing technologies offer a promising solution to this challenge. In this study, we adapt ChangeFormer, a transformer-based framework, to detect deforestation in the Brazilian Amazon, employing the attention mechanism to analyze spatial and temporal patterns in bitemporal satellite images. To assess the model's effectiveness, we employed a robust approach to create a deforestation detection (DD) dataset, utilizing Sentinel-2 imagery from select conservation areas in the Brazilian Amazon throughout 2020 and 2021. Our dataset comprises 7734 pairs of bitemporal image chips with a resolution of 256 times pixels and 1406 pairs of image chips with a resolution of 512 times pixels. The model achieved an overall accuracy (OA) of 93% with a corresponding F1 score of 90% and an intersection over union (IoU) score of 82%. These results demonstrate the potential of transformer-based networks for accurate and efficient DD.