Sentinel-1 SAR Based Weakly Supervised Learning For Tropical Forest Mapping

TL;DR

This paper introduces two weakly supervised learning methods using Sentinel-1 SAR data to accurately map tropical forests, reducing the need for extensive manual annotation.

Contribution

It presents novel weakly supervised approaches leveraging SAR data and sparse labels for tropical forest mapping, enabling high-quality results with less manual effort.

Findings

High quality forest maps achieved without manual annotations

Effective use of SAR data in weakly supervised learning

Applicable in tropical Amazon environment

Abstract

Tropical forests play an important role in regulating the global carbon cycle and are crucial for maintaining the tropical forest biodiversity. Therefore, there is an urgent need to map the extent of tropical forest ecosystems. Recently, deep learning has come out as a powerful tool to map these ecosystems with the caveat of curating high quality reference datasets. Since, manually annotating high quality reference datasets is time consuming and expensive, weakly supervised learning techniques offer the potential to train high quality models without the need for manually annotating large quantities of reference datasets. In this manuscript, we propose two weakly supervised approaches that are based on Sentinel-1 SAR images, sparsely distributed pixel-wise high quality reference labels and densely distributed noisy reference labels. The proposed approaches were tested in a tropical setting in the Brazilian amazon. The results demonstrate that high quality tropical forest maps can be derived from weakly supervised learning without the need for manually annotated labels.