Multitemporal classification without new labels: A solution with optimal transport

Tuia, Devis; Flamary, Rémi; Rakotomamonjy, Alain; Courty, Nicolas

doi:10.1109/multi-temp.2015.7245773

Cited by 5 publications

(5 citation statements)

References 12 publications

(11 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While the distribution estimation for the source domain can be done using the supervised classification, the estimation for the target domain is not trivial. A satisfying compromise is to use a uniform distribution [34], ensuring that it is respected when selecting samples. Sample selection must also ensure that an acceptable representation of each class is maintained.…”

Section: Problem Simplificationmentioning

confidence: 99%

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

2019

View full text Add to dashboard Cite

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%).

show abstract

Section: Problem Simplificationmentioning

confidence: 99%

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

2019

View full text Add to dashboard Cite

show abstract

“…Data set We consider the task of classifying superpixels from satellite images at very high resolution into a set of land cover/land use classes (Tuia et al , 2015). We use the 'Zurich Summer' data set…”

Section: Real-world Data From Remote Sensing Applicationmentioning

confidence: 99%

“…The algorithm proposed by the authors produces a set of hypotheses with one hypothesis per domain that can be combined using the theoretical study on multi-source domain adaptation presented in (Mansour et al , 2009b). Despite the rather small corpus of works in the literature dealing with the subject, target shift often occurs in practice, especially in applications dealing with anomaly/novelty detection (Blanchard et al , 2010;Scott et al , 2013;Sanderson & Scott, 2014), or in tasks where spatially located training sets are used to classify wider areas, as in remote sensing image classification (Tuia et al , 2015;Zhang et al , 2015).…”

Section: Introductionmentioning

confidence: 99%

Optimal Transport for Multi-source Domain Adaptation under Target Shift

Redko¹,

Courty²,

Flamary³

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

In this paper, we tackle the problem of reducing discrepancies between multiple domains, i.e. multi-source domain adaptation, and consider it under the target shift assumption: in all domains we aim to solve a classification problem with the same output classes, but with different labels proportions. This problem, generally ignored in the vast majority of domain adaptation papers, is nevertheless critical in real-world applications, and we theoretically show its impact on the success of the adaptation. Our proposed method is based on optimal transport, a theory that has been successfully used to tackle adaptation problems in machine learning. The introduced approach, Joint Class Proportion and Optimal Transport (JCPOT), performs multi-source adaptation and target shift correction simultaneously by learning the class probabilities of the unlabeled target sample and the coupling allowing to align two (or more) probability distributions. Experiments on both synthetic and real-world data (satellite image pixel classification) task show the superiority of the proposed method over the state-of-the-art.

show abstract

“…The compensation for such distortions, or shifts is one of the lively areas of machine learning, domain adaptation [17]. OT is providing a natural solution to this problem, by allowing to devise a non-linear transformation in the spectral dimensions of the image that helps in matching both distributions [18]. This transformation is directly obtained through the barycentric interpolation presented in Eq.…”

Section: Domain Adaptationmentioning

confidence: 99%

Optimal transport for data fusion in remote sensing

Courty

Flamary

Tuia

et al. 2016

2016 IEEE International Geoscience and Remote Sensing Symposium (IGARSS)

Self Cite

View full text Add to dashboard Cite

One of the main objective of data fusion is the integration of several acquisition of the same physical object, in order to build a new consistent representation that embeds all the information from the different modalities. In this paper, we propose the use of optimal transport theory as a powerful mean of establishing correspondences between the modalities. After reviewing important properties and computational aspects, we showcase its application to three remote sensing fusion problems: domain adaptation, time series averaging and change detection in LIDAR data. ABSTRACT One of the main objective of data fusion is the integration of several acquisition of the same physical object, in order to build a new consistent representation that embeds all the information from the different modalities. In this paper, we propose the use of optimal transport theory as a powerful mean of establishing correspondences between the modalities. After reviewing important properties and computational aspects, we showcase its application to three remote sensing fusion problems: domain adaptation, time series averaging and change detection in LIDAR data.

show abstract

Multitemporal classification without new labels: A solution with optimal transport

Cited by 5 publications

References 12 publications

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

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

Optimal Transport for Multi-source Domain Adaptation under Target Shift

Optimal transport for data fusion in remote sensing

Contact Info

Product

Resources

About