Assessment of Optimal Transport for Operational Land-Cover Mapping Using High-Resolution Satellite Images Time Series without Reference Data of the Mapping Period

Abstract

Land-cover map production using remote-sensing imagery is governed by data availability. In our case, data sources are two-fold: on one hand, optical data provided regularly by satellites such as Sentinel-2, and on the other hand, reference data which allow calibrating mapping methods or validating the results. The lengthy delays due to reference data collection and cleansing are one of the main issues for applications. In this work, the use of Optimal Transport (OT) is proposed. OT is a Domain Adaptation method that uses past data, both images and reference data, to produce the land-cover map of the current period without updated reference data. Seven years of Formosat-2 image time series and the corresponding reference data are used to evaluate two OT algorithms: conventional EMD transport and regularized transport based on the Sinkhorn distance. The contribution of OT to a classification fusion strategy is also evaluated. The results show that with a 17-class nomenclature the problem is too complex for the Sinkhorn algorithm, which provides maps with an Overall Accuracy (OA) of 30%. In contrast, with the EMD algorithm, an OA close to 70% is obtained. One limitation of OT is the number of classes that can be considered at the same time. Simplification schemes are proposed to reduce the number of classes to be transported. Cases of improvement are shown when the problem is simplified, with an improvement in OA varying from 5% and 20%, producing maps with an OA near 79%. As several years are available, the OT approaches are compared to standard fusion schemes, like majority voting. The gain in voting strategies with OT use is lower than the gain obtained with standard majority voting (around 5%).

Land Cover Maps Production with High Resolution Satellite Image Time Series and Convolutional Neural Networks: Adaptations and Limits for Operational Systems

“…In this context, the ability to fine-tune a pre-existing model is very useful. Indeed, climate differences between years don't allow to use a previoulsy trained model on newly acquired data [39]. While training a CNN model as FG-UNET for every map to be produced is too costly, fine-tuning the model has a cost comparable to the one of training a simpler classifier as RF.…”

ColorMapGAN: Unsupervised Domain Adaptation for Semantic Segmentation Using Color Mapping Generative Adversarial Networks

“…Recently, Courty et al have used the optimal transport for domain adaptation [56], [57]. The optimal transport for domain adaptation of remote sensing data is studied in [58]. Besides, after GANs had become popular in the field of computer vision, some papers have studied their application to remote sensing.…”

Vibration‐based detection and classification of structural changes using principal component analysis and ‐distributed stochastic neighbor embedding

“…With the information provided by the four actuation phases, we consider several approaches to finally classify the structure that must be diagnosed. One of these approaches, the majority voting, is widely used in standard fusion schemes, as well as weighted majority vote or soft voting . For the case studies in this work, the majority voting will be used, as well as an approach based on the sum of the inverse distances between the centroids and the map point, which is somehow related to a weighted majority vote.…”