Texture classification of photographic papers: improving spectral clustering using filterbanks on graphs

Nicolas Tremblay (1) , Stéphane G. Roux (1) , Pierre Borgnat (1) , Patrice Abry (2) , Herwig Wendt (3, 4) , Paul Messier (5)

1 Phys-ENS - Laboratoire de Physique de l'ENS Lyon
2 Paul Messier (USA)
3 IRIT-SC - Signal et Communications
4 CNRS - Centre National de la Recherche Scientifique
5 Chercheur indépendant

From the point of view of graph signal processing, we show that spectral clustering is equivalent to an ideal low-pass filterbank. Building upon previous multiscale community detection ideas [11], and integrating the concept of community cores [8], we propose a data-driven filterbank-based classification method. We apply this method to the texture classification of photographic papers useful to art historians, and we show that it provides a richer and more informative description of the data’s structure in clusters.

Mots clés: Filterbank; Print; Spectral clustering; Community detection; Art historian

Domaines: Synthèse d'image et réalité virtuelle [cs.GR]; Traitement du signal et de l'image [eess.SP]; Traitement des images [eess.IV]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Intelligence artificielle [cs.AI]

Nicolas Tremblay, Stéphane G. Roux, Pierre Borgnat, Patrice Abry, Herwig Wendt, et al.. Texture classification of photographic papers: improving spectral clustering using filterbanks on graphs. 25eme Colloque Groupe de Recherche et d'Etudes du Traitement du Signal et des Images (GRETSI 2015), Sep 2015, Lyon, France. pp. 1-4. ⟨hal-01518069⟩