Relative Spectral Difference Occurrence Matrix: A Metrological Spectral-Spatial Feature for Hyperspectral Texture Analysis - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2019

Relative Spectral Difference Occurrence Matrix: A Metrological Spectral-Spatial Feature for Hyperspectral Texture Analysis

Résumé

We develop a spectral-spatial feature, Relative Spectral Difference Occurrence Matrix (RSDOM) for hyperspectral texture recognition. Inspired by Julesz's conjecture, the proposed feature is based on spectral difference approach and respects the metro-logical constraints. It simultaneously considers the distribution of spectra and their spatial arrangement in the hyperspectral image. The feature is generic and adapted for any number of spectral bands and range. We validate our proposition by applying a classification scheme on the HyTexiLa database. An accuracy comparable to local binary pattern approach is obtained, but at a much reduced cost due to the extremely small feature size.
Fichier principal
Vignette du fichier
RSDOM.pdf (646.2 Ko) Télécharger le fichier
Loading...

Dates et versions

hal-02481614 , version 1 (21-02-2020)

Identifiants

Citer

Rui Jian Chu, Noël Richard, Christine Fernandez-Maloigne, Jon Yngve Hardeberg. Relative Spectral Difference Occurrence Matrix: A Metrological Spectral-Spatial Feature for Hyperspectral Texture Analysis. 27th Color and Imaging Conference Final Program and Proceedings, Oct 2019, Paris, France. pp.386-392, ⟨10.2352/issn.2169-2629.2019.27.69⟩. ⟨hal-02481614⟩
41 Consultations
106 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More