Randomized Nonlinear Component Analysis for Dimensionality Reduction of Hyperspectral Images

Bharath Damodaran 1 Nicolas Courty 1 Romain Tavenard 1
1 OBELIX - Environment observation with complex imagery
UBS - Université de Bretagne Sud, IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : Kernel based feature extraction method overcomes the curse of dimensionality and captures the non-linearities present in the data. However, these methods are not scal-able with large number of pixels found with hyperspec-tral images. Thus, a small subset of pixels are randomly selected to make the solution of kernel based methods tractable. In this paper, we propose scalable nonlinear component analysis for dimensionality reduction of hy-perspectral images. The proposed method relies on the randomized feature maps to capture the non-linearities between the variables in the hyperspectral data. Experiments conducted with three hyperspectral datasets show that our proposed method has provided better quality components and outperformed the state-of-the-art in terms of classification performance.
Liste complète des métadonnées

Cited literature [6 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01620604
Contributor : Nicolas Courty <>
Submitted on : Friday, October 20, 2017 - 5:26:10 PM
Last modification on : Monday, February 11, 2019 - 2:57:13 PM
Document(s) archivé(s) le : Sunday, January 21, 2018 - 1:45:02 PM

File

RNLCA.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01620604, version 1

Citation

Bharath Damodaran, Nicolas Courty, Romain Tavenard. Randomized Nonlinear Component Analysis for Dimensionality Reduction of Hyperspectral Images. IGARSS 2017 - IEEE International Geoscience and Remote Sensing Symposium, Jul 2017, Houston, United States. pp.1-4. ⟨hal-01620604⟩

Share

Metrics

Record views

544

Files downloads

339