Abstract
Abstract. Deforestation is a wide-reaching problem, responsible for serious environmental issues, such as biodiversity loss and global climate change. Containing approximately ten percent of all biomass on the planet and home to one tenth of the known species, the Amazon biome has faced important deforestation pressure in the last decades. Devising efficient deforestation detection methods is, therefore, key to combat illegal deforestation and to aid in the conception of public policies directed to promote sustainable development in the Amazon. In this work, we implement and evaluate a deforestation detection approach which is based on a Fully Convolutional, Deep Learning (DL) model: the DeepLabv3+. We compare the results obtained with the devised approach to those obtained with previously proposed DL-based methods (Early Fusion and Siamese Convolutional Network) using Landsat OLI-8 images acquired at different dates, covering a region of the Amazon forest. In order to evaluate the sensitivity of the methods to the amount of training data, we also evaluate them using varying training sample set sizes. The results show that all tested variants of the proposed method significantly outperform the other DL-based methods in terms of overall accuracy and F1-score. The gains in performance were even more substantial when limited amounts of samples were used in training the evaluated methods.
Highlights
Covering an area of approximately 5.5 million km2, which is equivalent to approximately one third the size of the South American continent, the Amazon rainforest encompasses half of the remaining tropical forest area on the planet (World Wildlife Fund, 2020a)
We evaluate an approach based on a specific Deep Learning (DL) Fully Convolutional Network (FCN) architecture, the DeepLabv3+ (Chen et al, 2018b), which we adapted to deforestation change detection
In this work we evaluated three deep learning-based methods employed for the task of deforestation detection in the Amazon rainforest
Summary
Covering an area of approximately 5.5 million km, which is equivalent to approximately one third the size of the South American continent, the Amazon rainforest encompasses half of the remaining tropical forest area on the planet (World Wildlife Fund, 2020a). Home to the largest collection of plants and animal species on the planet, the Amazon biome contains unparalleled biodiversity: it is the natural habitat of one tenth of the known species in the world (The Worldwatch Institute, 2015). The forest covers most of the Amazon river basin, source of 20% of all free-flowing fresh water on Earth (Assuncao, Rocha, 2019). The induced rain is responsible for irrigating crops and for filling the rivers and dams used to generate electrical energy by a large number of hydropower plants
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
