Weakly supervised alignment of multisensor images

Gonzalez, Diego Marcos; Camps‐Valls, Gustau; Tuia, Devis

doi:10.1109/igarss.2015.7326341

Cited by 5 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [40], the authors relax this requirement by working on semantic ties, i.e., samples issued from the same object but whose class is unknown. This last method therefore requires at least a partial overlap between the images to find the ties, either manually or by stereo matching, as in [41].…”

Section: Adapting Data Distributionsmentioning

confidence: 99%

“…Then the labeled pixels are projected into the target domain and are used to train a classifier therein. As for [40], partial overlap between the images is required. As one can see in Table 1, some methods will be more suitable than others, depending on the problem.…”

Section: Adapting Data Distributionsmentioning

confidence: 99%

“…However, it has the disadvantage of requiring labeled samples in all domains. In [40], the authors relax this requirement by working on semantic ties, i.e. samples issued from the same object, but whose class is unknown.…”

Section: B Adapting Data Distributionsmentioning

confidence: 99%

See 2 more Smart Citations

Domain Adaptation for the Classification of Remote Sensing Data: An Overview of Recent Advances

Tuia

Persello

Bruzzone

2016

IEEE Geosci. Remote Sens. Mag.

Self Cite

449

213

View full text Add to dashboard Cite

jUNE 2016 ieee Geoscience and remote sensinG maGazine 0274-6638/16©2016IEEE 41 T he success of the supervised classification of remotely sensed images acquired over large geographical areas or at short time intervals strongly depends on the representativity of the samples used to train the classification algorithm and to define the model. When training samples are collected from an image or a spatial region that is different from the one used for mapping, spectral shifts between the two distributions are likely to make the model fail. Such shifts are generally due to differences in acquisition and atmospheric conditions or to changes in the nature of the object observed. To design classification methods that are robust to data set shifts, recent remote sensing literature has considered solutions based on domain adaptation (DA) approaches. Inspired by machine-learning literature, several DA methods have been proposed to solve specific problems in remote sensing data classification. This article provides a critical review of the recent advances in DA approaches for remote sensing and presents an overview of DA methods divided into four categories: 1) invariant feature selection, 2) representation matching, 3) adaptation of classifiers, and 4) selective sampling. We provide an overview of recent

show abstract

Section: Adapting Data Distributionsmentioning

confidence: 99%

Section: Adapting Data Distributionsmentioning

confidence: 99%

See 1 more Smart Citation

Domain Adaptation for the Classification of Remote Sensing Data: An Overview of Recent Advances

Tuia

Persello

Bruzzone

2016

IEEE Geosci. Remote Sens. Mag.

Self Cite

449

213

View full text Add to dashboard Cite

show abstract

“…In the common feature space, the classifier trained by the source labeled data can perform well on the target data. The domain invariant features can be learned by minimizing the distribution differences between domains, where the data distributions are often described by the sample means [10]- [13], sample covariance matrix [14], [15], subspace eigenvectors [16], [17], or data manifold [18]- [21]. Lately, deep learning methods have been successfully applied for domain adaptation.…”

Section: Introductionmentioning

confidence: 99%

Combining Multiple Classifiers for Domain Adaptation of Remote Sensing Image Classification

Wei

Liu

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

“…Jun et al [19] proposed a method that separates the spatially varying component from the original spectral features and utilizes the residual information to model a Gaussian process maximum likelihood model (GP-ML). Manifold alignment techniques proposed in [20][21][22][23][24] also focus on projecting samples from both domains into a common space while preserving local manifold structures of the datasets during the transformation. In [25], the authors proposed a three-layer domain adaptation technique for a multi-temporal very high resolution (VHR) image classification problem.…”

Section: Introductionmentioning

confidence: 99%

Domain Adversarial Neural Networks for Large-Scale Land Cover Classification

Bejiga

Melgani

Beraldini³

2019

Remote Sensing

View full text Add to dashboard Cite

Learning classification models require sufficiently labeled training samples, however, collecting labeled samples for every new problem is time-consuming and costly. An alternative approach is to transfer knowledge from one problem to another, which is called transfer learning. Domain adaptation (DA) is a type of transfer learning that aims to find a new latent space where the domain discrepancy between the source and the target domain is negligible. In this work, we propose an unsupervised DA technique called domain adversarial neural networks (DANNs), composed of a feature extractor, a class predictor, and domain classifier blocks, for large-scale land cover classification. Contrary to the traditional methods that perform representation and classifier learning in separate stages, DANNs combine them into a single stage, thereby learning a new representation of the input data that is both domain-invariant and discriminative. Once trained, the classifier of a DANN can be used to predict both source and target domain labels. Additionally, we also modify the domain classifier of a DANN to evaluate its suitability for multi-target domain adaptation problems. Experimental results obtained for both single and multiple target DA problems show that the proposed method provides a performance gain of up to 40%.

show abstract

Weakly supervised alignment of multisensor images

Cited by 5 publications

References 12 publications

Domain Adaptation for the Classification of Remote Sensing Data: An Overview of Recent Advances

Domain Adaptation for the Classification of Remote Sensing Data: An Overview of Recent Advances

Combining Multiple Classifiers for Domain Adaptation of Remote Sensing Image Classification

Domain Adversarial Neural Networks for Large-Scale Land Cover Classification

Contact Info

Product

Resources

About