Graph Matching for Adaptation in Remote Sensing

Tuia, Devis; Muñoz-Marí, Jordi; Gómez-Chova, L.; Malo, Jesús

doi:10.1109/tgrs.2012.2200045

Cited by 90 publications

(69 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…by using feature selection (Gopalan et al, 2011) or feature extraction (Matasci et al, 2015). Some of the methods in this category are driven by a graph matching procedure to find correspondences between domains (Tuia et al, 2013;Banerjee et al, 2015). These methods need to contain the correct matching sequence among the possible matches or labelled samples across domains to perform well.…”

Section: Related Workmentioning

confidence: 99%

Boosted Unsupervised Multi-Source Selection for Domain Adaptation

Vogt

Paul

Östermann

et al. 2017

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Supervised machine learning needs high quality, densely sampled and labelled training data. Transfer learning (TL) techniques have been devised to reduce this dependency by adapting classifiers trained on different, but related, (source) training data to new (target) data sets. A problem in TL is how to quantify the relatedness of a source quickly and robustly, because transferring knowledge from unrelated data can degrade the performance of a classifier. In this paper, we propose a method that can select a nearly optimal source from a large number of candidate sources. This operation depends only on the marginal probability distributions of the data, thus allowing the use of the often abundant unlabelled data. We extend this method to multi-source selection by optimizing a weighted combination of sources. The source weights are computed using a very fast boosting-like optimization scheme. The run-time complexity of our method scales linearly in regard to the number of candidate sources and the size of the training set and is thus applicable to very large data sets. We also propose a modification of an existing TL algorithm to handle multiple weighted training sets. Our method is evaluated on five survey regions. The experiments show that our source selection method is effective in discriminating between related and unrelated sources, almost always generating results within 3% in overall accuracy of a classifier based on fully labelled training data. We also show that using the selected source as training data for a TL method will additionally result in a performance improvement.

show abstract

Section: Related Workmentioning

confidence: 99%

Boosted Unsupervised Multi-Source Selection for Domain Adaptation

Vogt

Paul

Östermann

et al. 2017

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…Once this mapping has been established, the feature samples from both domains can be transferred to the joint representation, thus allowing the application of the classifier trained on source data in the transformed domain without any adaptation. An unsupervised feature transfer method based on feature space clustering and graph matching is proposed in (Tuia et al, 2013). Experiments based on synthetic and real data show good results.…”

Section: Related Workmentioning

confidence: 99%

Iterative Re-Weighted Instance Transfer for Domain Adaptation

Paul¹,

Rottensteiner²,

Heipke³

2016

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

ABSTRACT:Domain adaptation techniques in transfer learning try to reduce the amount of training data required for classification by adapting a classifier trained on samples from a source domain to a new data set (target domain) where the features may have different distributions. In this paper, we propose a new technique for domain adaptation based on logistic regression. Starting with a classifier trained on training data from the source domain, we iteratively include target domain samples for which class labels have been obtained from the current state of the classifier, while at the same time removing source domain samples. In each iteration the classifier is re-trained, so that the decision boundaries are slowly transferred to the distribution of the target features. To make the transfer procedure more robust we introduce weights as a function of distance from the decision boundary and a new way of regularisation. Our methodology is evaluated using a benchmark data set consisting of aerial images and digital surface models. The experimental results show that in the majority of cases our domain adaptation approach can lead to an improvement of the classification accuracy without additional training data, but also indicate remaining problems if the difference in the feature distributions becomes too large.

show abstract

“…10). Model portability has been considered by many researchers for analysis of data over extended spatial areas and slowly varying multi-temporal scenarios using various approaches, including signature extension through clustering [58], spatially invariant features [59] obtained by spatial detrending with Gaussian processes in [60], and by manifold alignment [61]. The resulting classifiers were more robust to local shift in areas where training samples were unavailable.…”

Section: Active Learning For Knowledge Transfer and Adaptation Promentioning

confidence: 99%

Active Learning: Any Value for Classification of Remotely Sensed Data?

2013

Self Cite

View full text Add to dashboard Cite

Active learning, which has a strong impact on processing data prior to the classification phase, is an active research area within the machine learning community, and is now being extended for remote sensing applications. To be effective, classification must rely on the most informative pixels, while the training set should be as compact as possible. Active learning heuristics provide capability to select unlabeled data that are the "most informative" and to obtain the respective labels, contributing to both goals. Characteristics of remotely sensed image data provide both challenges and opportunities to exploit the potential advantages of active learning. We provide an overview of active learning methods, then review the latest techniques proposed to cope with the problem of interactive sampling of training pixels for classification of remotely sensed data with support vector machines. We discuss remote sensing specific approaches dealing with multi-source and spatially and time-varying data, and provide examples for high dimensional hyperspectral and very high resolution multispectral imagery.

show abstract

Graph Matching for Adaptation in Remote Sensing

Cited by 90 publications

References 52 publications

Boosted Unsupervised Multi-Source Selection for Domain Adaptation

Boosted Unsupervised Multi-Source Selection for Domain Adaptation

Iterative Re-Weighted Instance Transfer for Domain Adaptation

Active Learning: Any Value for Classification of Remotely Sensed Data?

Contact Info

Product

Resources

About