INTELLIGENT GEO-SEARCH SYSTEM: A SEMANTIC AWARE METHOD FOR RETRIEVING REMOTE SENSING IMAGES Dimitris Sykas 1 , Alkyoni Baglatzi 2 , George Vafiadis 3 , Vrettos Moulos 3 , Vijay Kumar Vohora 4 1 Laboratory of Remote Sensing, National Technical University of Athens H. Polytechniou Str. 9, 15780 Zografos Campus, Greece dimsyk@gmail.com 2 Laboratory of Cartography, National Technical University of Athens H. Polytechniou Str. 9, 15780 Zografos Campus, Greece baglatzi@mail.ntua.gr 3 MapTronic Limited 2 Woodberry Grove, London N12 0DR, UK, {gvaiadis, vrettos}@maptronic.co.uk 4 Infotech Enterprises Europe Limited 52-54 High Holborn, London WC1V 6RL, United Kingdom, v.vohora@gmail.com Abstract: Content-based remote sensing image retrieval techniques have been introduced for bridging the gap between low-level semantics of images and high-level semantics of user queries. The proposed Intelligent Geo-Search System (IGSS) is a knowledge aware, spectral oriented retrieval methodology. Knowledge about geographic objects and processes is formalized as an ontology. A reference spectral library is built consisting of spectral signatures. Tags are assigned to images using an endmember extraction algorithm and a labeling algorithm. Indexes such as (NDVI, NDMI, NBRI) and additional statistics for each index are stored along with the tags. In that way, queries can be formulated that enable both geographic entities detection (e.g. burned areas and forest type identification) and phenomena quantification (e.g. increased risk for forest fire), enabling more robust domain oriented question answering. For demonstration purposes Landsat 7 ETM images in the vegetation domain have been chosen. The results demonstrate that images can be retrieved with 89.1% spectral matching accuracy. Keywords: tagging, semantic search, endmembers, ontologies, image retrieval 1. INTRODUCTION Technological advances in remote sensing increased the availability of satellite images with different spatiotemporal and spectral characteristics. The problem of obtaining data has been trans- formed into the difficulty of retrieving the most appropriate data for each user's needs. The biggest challenge is to bridge the gap between the low-level semantics (detectable, quantifiable features) of the images and the semantic information in them (Sethi et al., 2001) with the view to designing intelligent geo-search systems. This bottleneck between low-level semantics of images and high-level semantics of user quer- ies has already been acknowledged in the literature. Under the general term of content-based image re- trieval a wealth of approaches has been introduced. We point the reader to Liu et al., (2007), Rani and Reddy, (2012) and Zhang et al., (2012) for a review and comparison of approaches via the prism of computer science. Remote sensing images have special characteristics and therefore more specialized methodo- logies are needed (Xiran et al., 2012). Related work spans from the use of different data i.e. Vegan- zones et al., (2008) focus on techniques for hyperspectral remote sensing images while in the EU-fun-