Sensitivity of Support Vector Machines to Random Feature Selection in Classification of Hyperspectral Data

By Björn Waske, Sebastian van der Linden, Jon Atli Benediktsson, Andreas Rabe, Patrick Hostert. IEEE Transactions on Geoscience and Remote Sensing, 2010

308 citations recorded by the source corpus.

Subfields: Atmospheric Science, Ecology, Media Technology

View Paper: https://www.semanticscholar.org/paper/3c07fc05471fbaeb2a216126ffeef6810e246cbe

How researchers see this paper

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

In Atmospheric Science

In Ecology

In Media Technology

Citations in the corpus

Feature Mining for Hyperspectral Image Classification (2013, background): “Improved performances are observed in [7] and [8].”
Deep learning classifiers for hyperspectral imaging: A review (2019, methodology): “Also, some of the most popular classification algorithms available in the literature have been considered: (1) random forest (RF), (2) multinomial logistic regression (MLR), (3) support vector machine (SVM) with radial basis function kernel (Waske et al., 2010), (4) multilayer perceptron (MLP), (5) ”
Deep Pyramidal Residual Networks for Spectral–Spatial Hyperspectral Image Classification (2019, methodology): “The proposed approach has been compared to a total of 10 different classification methods available in the literature: 1) SVM with radial basis function kernel [65]; 2) RF; 3) MLP; 4) extreme learning machine (ELM) [66]; 5) kernel ELM (KELM) [67]; 6) 1-D CNN; 7) 2-D CNN; 8) 3-D-CNN; 9) spectral–spat”
Diverse Region-Based CNN for Hyperspectral Image Classification (2018, methodology): “…with some state-of-the-art HSI classiﬁcation approaches, such as SVM with radial basis function (denoted as SVM-RBF) and multiple classiﬁer systems based SVM with random feature selection (denoted as SVM-RFS [50]), SVM-MRF [21], CNN [33], R-PCA CNN [35], CNN-PPF [38], CD-CNN [39], and SS-CNN [36].”
Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers (2014, background): “A number of authors have found that the radial basis function outperforms the polynomial function (Kavzoglu and Colkesen 2009b; Waske et al. 2010; Huang et al. 2008; Oommen et al. 2008).”
Capsule Networks for Hyperspectral Image Classification (2019, methodology): “Specifically, the SVM with radial basis function kernel [55],”
A Pixel-Based Landsat Compositing Algorithm for Large Area Land Cover Mapping (2013, methodology): “The RF models were trained using 500 individual trees and the number of features at each split was set to the square root of the number of input features [50].”
When a Tree Falls: Using Diversity in Ensemble Classifiers to Identify Evasion in Malware Detectors (2016, background): “Recent work has demonstrated that the diversity in ensemble classifiers can improve malware detection rates [22], [29], [46], [47].”
An SVM Ensemble Approach Combining Spectral, Structural, and Semantic Features for the Classification of High-Resolution Remotely Sensed Imagery (2013, background): “applied to classification of hyperspectral images [3], [37], multisource data [38], and high-resolution urban images [39].”
A Kernel-Based Feature Selection Method for SVM With RBF Kernel for Hyperspectral Image Classification (2014, background): “[52] or the distribution (importance) of feature weights [51]”
Flood Detection and Susceptibility Mapping Using Sentinel-1 Remote Sensing Data and a Machine Learning Approach: Hybrid Intelligence of Bagging Ensemble Based on K-Nearest Neighbor Classifier (2020, background): “Hm are termed ‘weak classifiers’ because their accuracy is higher than the accuracy of other random classifiers [119].”