GVIP Journal    

GVIP
 VOLUME={08}, ISSUE = {2} ICGST

Satellite Image Retrieval Based On Ontology Merging

Imed Riadh Farah1, 2, Wassim Messaoudi1,2, Karim saheb ettabâa1, 2 and Basel Solaiman2
1Laboratoire de Recherche en Informatique Arabisée et Documentique Intégrée (R.I.AD.I),
Ecole Nationale des Sciences de l’Informatique. Campus Universitaire de Manouba, 2010 Manouba, Tunis, Tunisie
2ENST-Bretagne, laboratoire ITI Technopôle Brest Iroise CS 83818, 29238 Brest Cedex France


Abstract

With the rapid development of remote sensing platform and sensor technique, the amount of satellite images has frequently increased. In order to analyze, manage and retrieve spatial information, several techniques are used to improve the quality of retrieval systems and to perform semantic in the retrieval process. In this paper, we propose an ontology-based approach for semantic retrieving of satellite images, describing the semantic content image and managing uncertain information. The proposed system is composed of three modules: ontological modeling of scene, ontological models merging and semantic retrieval. The first module, describes the semantic image content by an ontological model based on sensor model, scene model and spatial relations model. The second module develops a reliable ontological model of the satellite scene for merging incompletes models and managing uncertain information and conflict situations. The third module retrieves similar satellite images basing on their ontological models retrieves similar satellites images based on their ontological models.

 Keywords: Content-based image retrieval, scene interpretation, knowledge representation, ontology merging.

(P1150803001, 736 KB), (Presentation 2.65 MB)

Biographies:

Imed Riadh Farah, received the M.D. and Dr Eng. degrees, from ISG Inst. of computer sciences in 1995 and ENSI Sch, in 2003, respectively. After working as a research assistant (from 1996) and a permanent researcher at laboratory RIADI, ENSI National School of Computer Sciences engineering (since 1995), he has been an assistant professor at University of Jendouba, since 2004. His research interest includes image processing, pattern recognition, artificial intelligence, data mining and their application to remote sensing. He is a member of Arts-Pi Tunisia.

 

Wassim Messaoudi, received the Master engineering, from FSJEGJ in 2007. He is a permanent researcher at laboratory RIADI, ENSI National School of Computer Sciences engineering (since 2006). His research interest includes image processing, image fusion, artificial intelligence, data mining, pattern recognition and their application to remote sensing.

 

Karim Saheb Ettabaa, received the M. E. and Dr Eng, from ENSI in 2004 and 2007, respectively. He is a permanent researcher at laboratory RIADI, ENSI National School of Computer Sciences engineering (since 2004). His research interest includes image processing, data mining, artificial intelligence, pattern recognition and their application to remote sensing.

 

Bassel Solaiman, received the telecommunication engineering degree from Ecole Nationale Supérieure des télécommunications de Bretagne (Télécom Bretagne). Brest, France, in 1983 and the PhD degree from the University de Rennes I, Brest, France, in 1989. From 1984 to 1985, he was a Research Assistant in the Communication Group at the Centre d’Etudes et de recherche Scientifique, Damascus, Syria. He joined the Image and Information Processing Department at Télécom Bretagne, Brest, France, in 1992. His current research interests include the fields of remote sensing, medical image processing, pattern recognition, neural network, and artificial intelligence.

BibTex:

@ARTICLE{P1150803001,

AUTHOR = {Imed Riadh Farah and Wassim Messaoudi and Karim saheb ettabâa and Basel Solaiman},

TITLE = {Satellite Image Retrieval Based On Ontology Merging},

JOURNAL ={ICGST International Journal on Graphics, Vision and Image Processing, GVIP},

YEAR = {2008},

VOLUME = {08},

ISSUE ={II},

PAGES={45--53}

}

(P1150803001, 736 KB), (Presentation 2.65 MB)