A Novel Transductive SVM for Semisupervised Classification of Remote-Sensing Images

By Lorenzo Bruzzone, M. Chi, Mattia Marconcini. IEEE Transactions on Geoscience and Remote Sensing, 2006

599 citations recorded by the source corpus.

Subfields: Analytical Chemistry, Media Technology

View Paper: https://www.semanticscholar.org/paper/3e32b4775f3aeff0e1eaf37cc46771376a0c8d08

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Analytical Chemistry

In Media Technology

Citations in the corpus

Spectral and Spatial Classification of Hyperspectral Data Using SVMs and Morphological Profiles (2008, methodology): “Semisupervised SVM were also investigated for multi-spectral data classiﬁcation [27], [28].”
Semi-Supervised Graph-Based Hyperspectral Image Classification (2007, methodology): “2) Low-density separation algorithms which maximize the margin for labeled and unlabeled samples simultaneously, such as transductive SVM [25], which have been recently applied to hyperspectral image classification [ 26 ].”
Semi-Supervised Deep Learning Using Pseudo Labels for Hyperspectral Image Classification (2018, methodology): “TSVM exploits speciﬁc iterative algorithms that gradually search a reliable separating hyperplane (in the kernel space) with a transductive process that incorporates both labeled and unlabeled samples in the training phase.”
Potential, challenges and future directions for deep learning in prognostics and health management applications (2020, background): “These include (deep) generative models (Kingma et al., 0000), graph-based methods (Zhao et al., 2015) and transductive methods (Bruzzone et al., 2006), to name a few.”
Support vector machines in remote sensing: A review (2011, unknown): “RVM incorporatedpriorknowledgeoftheproblemandprovedtobeuse-ful in quantifying chlorophyll concentrations based on ocean surface reflectance.”
Domain Adaptation for the Classification of Remote Sensing Data: An Overview of Recent Advances (2016, methodology): “In [58], the authors employ the trans-ductive SVM, a method that iteratively moves the decision boundary of the SVM classifier toward low-density areas of the unlabeled target domain.”
Land-cover classification with high-resolution remote sensing images using transferable deep models (2018, background): “They are commonly based on the selection of invariant features (Izquierdo-Verdiguier et al., 2013; Bruzzone and Persello, 2009), the adaptation of data distributions (Tuia and Camps-Valls, 2016; Yang and Crawford, 2016; Jun and Ghosh, 2011), and active learning (Persello and Bruzzone, 2012; Demir…”
Support Vector Machine Versus Random Forest for Remote Sensing Image Classification: A Meta-Analysis and Systematic Review (2020, unknown): “often delivering higher classification accuracy compared to the conventional methods [95]–[98].”
Semisupervised Classification of Remote Sensing Images With Active Queries (2012, background): “approaches exist in the remote sensing literature [8]–[11].”
Generalized Composite Kernel Framework for Hyperspectral Image Classification (2013, background): “On the other hand, the collection of reliable training samples is very expensive in terms of time and finance, and the possibility to exploit large ground truth information is not common [4].”
Mean Map Kernel Methods for Semisupervised Cloud Classification (2010, background): “classification either based on transductive approaches, graphs, or Laplacian SVMs [22]–[25].”
Monitoring Land-Cover Changes: A Machine-Learning Perspective (2016, unknown): “[31] L.”
Semisupervised Neural Networks for Efficient Hyperspectral Image Classification (2010, background): “The TSVM, originally proposed in [9] and further extended to deal with the peculiarities of remote sensing data in [10], aims at choosing a decision boundary that maximizes the margin on both labeled and unlabeled data.”
Recent Advances in Techniques for Hyperspectral Image Processing (2009, methodology): “Among the great many methods proposed in the literature for such approaches (Dempster77, Chapelle06), we focus on the transductive SVM for semi-supervised learning (Bruzzone et al., 2006), and in the composite kernels methodology (Camps-Valls et al., 2006) for contextual information integration.”
Advances in Hyperspectral Image Classification: Earth Monitoring with Statistical Learning Methods (2013, background): “Low-density TSVM [11] Look for the emptiest margin”
Urban-Area and Building Detection Using SIFT Keypoints and Graph Theory (2009, background): “Some related papers can be counted as [4], [6], [10], [15]–[17], and [20].”