Temporal-Domain Adaptation for Satellite Image Time-Series Land-Cover Mapping With Adversarial Learning and Spatially Aware Self-Training

Abstract

Listen

Translate article

Nowadays, Satellite Image Time Series (SITS) are commonly employed to derive land cover maps (LCM) to support decision makers in a variety of land management applications. In the most general workflow, the production of LCM strongly relies on available ground truth data to train supervised machine learning models. Unfortunately, this data is not always available due to time-consuming and costly field campaigns. In this scenario, the possibility to transfer a model learnt on a particular year ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">source domain</i> ) to a successive period of time ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">target domain</i> ), over the same study area, can save time and money. Such a kind of model transfer is challenging due to different acquisition conditions affecting each time period thus, resulting in possible distribution shifts between <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">source</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">target</i> domains. In the general field of machine learning, Unsupervised Domain Adaptation (UDA) approaches are well suited to cope with the learning of models under distribution shifts between <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">source</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">target</i> domains. While widely explored in the general computer vision field, they are still under investigated for SITS-based land cover mapping, especially for the temporal transfer scenario. With the aim to cope with this scenario in the context of SITS-based land cover mapping, here we propose Spatially Aligned Domain-Adversarial Neural Network, a framework that combines both adversarial learning and self-training to transfer a classification model from a time period (year) to a successive one on a specific study area. Experimental assessment on a study area located in Burkina Faso characterized by challenging operational constraints demonstrates the significance of our proposal. The obtained results have shown that our proposal outperforms all the UDA competing methods by 7 to 12 points of F1-score across three different transfer tasks.