Deep learning tools to support deforestation monitoring in Ivory Coast using SAR and optical satellite imagery

Abstract

Deforestation is gaining increasing importance due to its strong influence on the surrounding environment, especially in developing countries where the population has a disadvantaged economic condition and agriculture is the main source of income. In Ivory Coast, for instance, where the cocoa production is the most remunerative activity, it is not rare to assist the replacement of portions of ancient forests with new cocoa plantations. To monitor this type of deleterious activity, satellites can be employed to recognize the disappearance of the forest. In this study, Forest-Non-Forest map (FNF) has been refined (from 25m/px to 10m/px) to be used as target for models based on Sentinel images input. State-of-the-art models U-Net, Attention U-Net, Segnet and FCN32 are compared over different years, combining Sentinel-1, Sentinel-2, and cloud probability to create forest/non-forest segmentation. Although Ivory Coast lacks of local forest coverage datasets and is partially covered by Sentinel images, it is demonstrated the feasibility of creating models classifying forest and non-forest pixels over the area using open datasets to predict where deforestation could have occurred. Although a significant portion of the deforestation research is carried out on visible bands, SAR acquisitions and the cloud probability layer are employed to overcome the limits of RGB images over areas often covered by clouds. Finally, the most promising models are employed to estimate the forest that has been cut between 2019 and 2020.