Skip to Main content Skip to Navigation
Conference papers

Satellite image retrieval with pattern spectra descriptors

Petra Bosilj 1, 2 Erchan Aptoula 3 Sébastien Lefèvre 2 Ewa Kijak 1
1 LinkMedia - Creating and exploiting explicit links between multimedia fragments
Inria Rennes – Bretagne Atlantique , IRISA-D6 - MEDIA ET INTERACTIONS
2 OBELIX - Environment observation with complex imagery
UBS - Université de Bretagne Sud, IRISA-D5 - SIGNAUX ET IMAGES NUMÉRIQUES, ROBOTIQUE
Abstract : The increasing volume of Earth Observation data calls for appropriate solutions in satellite image retrieval. We address this problem by considering morphological descrip-tors called pattern spectra. Such descriptors are histogram-like structures that contain the information on the distribution of predefined properties (attributes) of image components. They can be computed both at the local and global scale, and are computationally attractive. We demonstrate how they can be embedded in an image retrieval framework and report their promising performances when dealing with a standard satellite image dataset.
Complete list of metadatas

Cited literature [25 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01253815
Contributor : Sébastien Lefèvre <>
Submitted on : Wednesday, November 13, 2019 - 6:40:00 PM
Last modification on : Thursday, April 2, 2020 - 1:54:48 AM

File

iim2015.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01253815, version 1

Citation

Petra Bosilj, Erchan Aptoula, Sébastien Lefèvre, Ewa Kijak. Satellite image retrieval with pattern spectra descriptors. International Conference on Image Information Mining: Geospatial Intelligence from Earth Observation (IIM), 2015, Bucarest, Romania. ⟨hal-01253815⟩

Share

Metrics

Record views

1245

Files downloads

21