Local 2D Pattern Spectra as Connected Region Descriptors

Petra Bosilj (1, 2) , Michael H.F. Wilkinson (3) , Ewa Kijak (1) , Sébastien Lefèvre (2)

1 LinkMedia - Creating and exploiting explicit links between multimedia fragments
2 OBELIX - Environment observation with complex imagery
3 Bernoulli Institute for Mathematics and Computer Science and Artificial Intelligence

We validate the usage of augmented 2D shape-size pattern spectra, calculated on arbitrary connected regions. The evaluation is performed on MSER regions and competitive performance with SIFT descriptors achieved in a simple retrieval system, by combining the local pattern spectra with normalized central moments. An additional advantage of the proposed descriptors is their size: being half the size of SIFT, they can handle larger databases in a time-efficient manner. We focus in this paper on presenting the challenges faced when transitioning from global pattern spectra to the local ones. An exhaustive study on the parameters and the properties of the newly constructed descriptor is the main contribution offered. We also consider possible improvements to the quality and computation efficiency of the proposed local descriptors.

Domaines: Traitement des images [eess.IV]

Petra Bosilj, Michael H.F. Wilkinson, Ewa Kijak, Sébastien Lefèvre. Local 2D Pattern Spectra as Connected Region Descriptors. International Symposium on Mathematical Morphology, 2015, Reykjavik, Iceland. pp.182-193, ⟨10.1007/978-3-319-18720-4_16⟩. ⟨hal-01168146⟩