Co-occurrence Matrix of Covariance Matrices: A Novel Coding Model for the Classification of Texture Images

Ioana Ilea 1, 2 Lionel Bombrun 2 Salem Said 2 Yannick Berthoumieu 2
1 UTCN - Universitatea Tehnica din Cluj-Napoca
2 IMS - Laboratoire de l'intégration, du matériau au système
Abstract : This paper introduces a novel local model for the classification of covariance matrices: the co-occurrence matrix of covariance matrices. Contrary to state-of-the-art models (BoRW, R-VLAD and RFV), this local model exploits the spatial distribution of the patches. Starting from the generative mixture model of Riemannian Gaussian distributions , we introduce this local model. An experiment on texture image classification is then conducted on the VisTex and Outex_TC000_13 databases to evaluate its potential.
Keywords : Co-occurrence matrix Riemannian Gaussian distributions classification covariance matrix
Document type :
Conference papers
Complete list of metadatas

Cited literature [21 references]  Display  Hide  Download
https://hal.archives-ouvertes.fr/hal-01629171
Contributor : Lionel Bombrun <>
Submitted on : Monday, November 6, 2017 - 10:12:56 AM
Last modification on : Thursday, February 7, 2019 - 4:48:03 PM

File

Ilea17_GSI.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01629171, version 1

Collections

CNRS | IMS-BORDEAUX-FUSION | IMS-BORDEAUX

Citation

Ioana Ilea, Lionel Bombrun, Salem Said, Yannick Berthoumieu. Co-occurrence Matrix of Covariance Matrices: A Novel Coding Model for the Classification of Texture Images. 3rd conference on Geometric Science of Information, Nov 2017, Paris, France. ⟨hal-01629171⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

92

Files downloads

257

Contact

CCSD