Support vector machines for hyperspectral image classification with spectral-based kernels

By Grégoire Mercier, Marc Lennon. IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477), 2003

311 citations recorded by the source corpus.

Subfields: Analytical Chemistry, Atmospheric Science, Media Technology

View Paper: https://www.semanticscholar.org/paper/7b22587b518d406e81e10c74719696e439fef2a4

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

1. Jon Atli Benediktsson
2. Jocelyn Chanussot
3. Gustau Camps-Valls
4. Antonio J. Plaza
5. Lorenzo Bruzzone
6. Luis Gómez-Chova
7. Mathieu Fauvel
8. Tatyana V. Bandos Marsheva

In Atmospheric Science

1. Jon Atli Benediktsson
2. Jocelyn Chanussot
3. Gustau Camps-Valls
4. Antonio J. Plaza
5. Lorenzo Bruzzone
6. Shutao Li
7. Mathieu Fauvel
8. Pedram Ghamisi

In Media Technology

1. Jon Atli Benediktsson
2. Antonio J. Plaza
3. Jocelyn Chanussot
4. Gustau Camps-Valls
5. Lorenzo Bruzzone
6. Shutao Li
7. Liangpei Zhang
8. Jun Yu Li

Citations in the corpus

• Composite kernels for hyperspectral image classification (2006, methodology): “Our immediate future work is tied to the use of other kernel distances, such as the Spectral Angle Mapper [24], and more sophisticated texture techniques for describing the spatial structure of the classes, such as Gabor filters, Markov random fields, and cooccurrence matrices [25].”
• Multiple support vector machines for land cover change detection: An application for mapping urban extensions (2006, methodology): “Hence, test data are classified according to: xa positive class if f ðxÞN0 negative class if f ðxÞb0 ð2Þ Furthermore, all mathematical functions, which satisfy Mercer's conditions, are eligible to be a SVMkernel (Mercier and Lennon, 2003).”
• Recent Advances in Techniques for Hyperspectral Image Processing (2009, methodology): “(Mercier & Lennon, 2003) defined a spectral angle kernel kSAM(x,z)= exp(−γα(x,z)2), α x; zð Þ = arccos hx;zijjx jj : jjz jj using the scale invariance property of the spectral data.”
• A relative evaluation of multiclass image classification by support vector machines (2004, background): “Key attractions of the SVM-based approach to classification are that it seeks to fit an optimal hyperplane between classes and may require only a small training sample [15], [18], [19].”
• Semi-Supervised Graph-Based Hyperspectral Image Classification (2007, methodology): “Next steps will consider the use of other kernel distances, such as the spectral angle mapper [51], and more sophisticated texture techniques for describing the spatial structure of the classes, such as Gabor filters, co-occurrence matrices [52], [53] or Markov Random Fields [54], [55].”
• Support vector machine-based arrhythmia classification using reduced features of heart rate variability signal (2008, background): “In this way, not only an optimal hyperplane is fitted, but also less training samples are effectively used; thus high classification accuracy is achieved with small training sets [49].”