Share Email Print
cover

Proceedings Paper

Scattering transforms and classification of hyperspectral images
Author(s): Wojciech Czaja; Ilya Kavalerov; Weilin Li
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

We explore the representation capabilities of scattering transforms for the classification of hyperspectral images. We examine several types, including a recently developed technique called the Fourier scattering transform. This method is naturally suited for the representation of hyperspectral data because it decomposes signals into multi-frequency bands and removes small perturbations such as noise. We test on four standard hyperspectral datasets, and the results indicate that the Fourier scattering transform is effective at representing spectral data. We also present a spatial-spectral scattering transform that combines Wavelet and Fourier representations, and this method obtains significantly higher classification accuracies.

Paper Details

Date Published: 8 May 2018
PDF: 12 pages
Proc. SPIE 10644, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXIV, 106440H (8 May 2018); doi: 10.1117/12.2305152
Show Author Affiliations
Wojciech Czaja, Univ. of Maryland, College Park (United States)
Ilya Kavalerov, Univ. of Maryland, College Park (United States)
Weilin Li, Univ. of Maryland, College Park (United States)


Published in SPIE Proceedings Vol. 10644:
Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XXIV
Miguel Velez-Reyes; David W. Messinger, Editor(s)

© SPIE. Terms of Use
Back to Top