Semi-Supervised Graph-Based Hyperspectral Image Classification

By Gustau Camps-Valls, Tatyana V. Bandos Marsheva, Dengyong Zhou. IEEE Transactions on Geoscience and Remote Sensing, 2007

613 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/7c997a648976fb9df320c6ca332c9dc155820454

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Batch-Mode Active-Learning Methods for the Interactive Classification of Remote Sensing Images (2011, background): “The semisupervised approach has been widely investigated in the recent years in the RS community [1]-[5].”
CNN-Enhanced Graph Convolutional Network With Pixel- and Superpixel-Level Feature Fusion for Hyperspectral Image Classification (2020, background): “Besides, the spatial structures of HSIs have also been explored to boost the spectral–spatial feature learning, such as graph construction [17], [18], superpixel segmentation [19], and morphological segmentation [20], [21].”
Semisupervised Neural Networks for Efficient Hyperspectral Image Classification (2010, background): “Recently, more attention has been paid to discriminative approaches, such as 1) the transductive SVM (TSVM) [9], [10], which simultaneously maximizes the margin for labeled and unlabeled samples; 2) graph-based methods, in which each pixel spreads its label information to its neighbors until a globa”
Revisiting Graph based Collaborative Filtering: A Linear Residual Graph Convolutional Network Approach (2020, neutral): “These models show competing performance for tasks such as unsupervised node (graph) representation learning (Li et al. 2016), semi-supervised node (graph) learning (Kipf and Welling 2017; Camps-Valls, Marsheva, and Zhou 2007), and so on.”
Kernel-Based Framework for Multitemporal and Multisource Remote Sensing Data Classification and Change Detection (2008, methodology): “At present, good results have been obtained by using graph-based methods [39]; however, it is worth stressing that any kernel-based transductive or one-class learning algorithm could be adapted for this purpose.”
Semi-supervised hyperspectral image segmentation (2009, methodology): “The results are compared with a semi-supervised classification approach [11], which is equivalent to our method without spatial prior, and with the semi-supervised method presented in [15].”
Learning to Diversify Deep Belief Networks for Hyperspectral Image Classification (2017, methodology): “Research in the last decades has resulted in many popular models, such as neural networks [7], [8], the Gaussian process approach [9], conditional random fields [10], [11], the graph method [12], [13], multinomial logistic regression [14], [15], support vector machines (SVMs) [16], AdaBoost [17], an”
Recent Advances in Techniques for Hyperspectral Image Processing (2009, background): “…withd={1,…10} formodeling the spectral features according toprevious results in the literature (Gualtieri et al., 1999; Camps-Valls & Bruzzone, 2005; Camps-Valls et al., 2007), and the Gaussian RBF kernel (with σ= {10−1,…,103}) for modeling the spatial features, according to the locality…”
SVM-Based Fuzzy Decision Trees for Classification of High Spatial Resolution Remote Sensing Images (2012, methodology): “Textural information is introduced by considering simple kernels according to the vector stacking approach [10] or composite kernel forms [11].”