In this paper, we demonstrate the concepts of a prototype of a knowledge-driven content-based information mining system produced to manage and explore large volumes of remote sensing image data. The system consists of a computationally intensive offline part and an online interface. The offline part aims at the extraction of primitive image features, their compression, and data reduction, the generation of a completely unsupervised image content-index, and the ingestion of the catalogue entry in the database management system. Then, the user's interests-semantic interpretations of the image content-are linked with Bayesian networks to the content-index. Since this calculation is only based on a few training samples, the link can be computed online, and the complete image archive can be searched for images that contain the defined cover type. Practical applications exemplified with different remote sensing datasets show the potential of the system. Index Terms-Content-based image retrieval (CBIR), image information mining, information extraction, statistical learning.
During the last decade, the exponential increase of multimedia and remote sensing image archives, the fast expansion of the world wide web, and the high diversity of users have yielded concepts and systems for successful content-based image retrieval and image information mining. Image data information systems require both database and visual capabilities, but there is a gap between these systems. Database systems usually do not deal with multidimensional pictorial structures and vision systems do not provide database query functions. In terms of these points, the evaluation of content-based image retrieval systems became a focus of research interest. One can find several system evaluation approaches in literature, however, only few of them go beyond precision-recall graphs and do not allow a detailed evaluation of an interactive image retrieval system. Apart from the existing evaluation methodologies, we aim at the overall validation of our knowledge-driven content-based image information mining system. In this paper, an evaluation approach is demonstrated that is based on information-theoretic quantities to determine the information flow between system levels of different semantic abstraction and to analyze human-computer interactions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.