OSIRIDE is a Content Based Image Retrieval platform for Earth Observation data archives.
With the current emergence of very large-scale image collections, the difficulties faced by the user based analysis and manual annotation approach are becoming more and more severe. To overcome them, OSIRIDE proposes techniques to index Earth Observation imagery by their own visual content, such as color and texture.
OSIRIDE has established a general framework of image retrieval from a new perspective, where automatic identification of relevant data based on semantic content is a key element in view of full exploitation of next generation of satellite images.
Funded by the European Space Agency in the frame of the General Support Technology Program.
Developed by leading Romanian actors in the field of Earth Observation: Terrasigna S.R.L and Research Center for Spatial Information University Politehnica of Bucharest
For reference please refer to the following publications:
C. Cucu-Dumitrescu, F. Serban, A. Stoica, C. Vaduva, M. Iapaolo: “OSIRIDE: Prototype system for content based image retrieval in satellite imagery”, 2016 Conference on Big Data from Space (BiDS) 15 - 17 March 2016, Santa Cruz de Tenerife, Spain, pp 336-339, doi: 10.2788/854791