An Information Retrieval Approach for Image Databases

The retrieval process in image database systems is inherently different from the retrieval process in traditional (record oriented) database systems. While the latter can be considered an exact process (records either satisfy the query or do not so), the former is not an exact process and the system must take into account the uncertainty factor (i.e. the answer is not only "true" or "false" but is often in between them). Uncertainty is mainly introduced during the image analysis process (i.e. when semantic objects are associated, with a certain recognition degree, to image components, and multiple image interpretations are generated) and in the evaluation of how complex objects in images are relevant to user's queries. Moreover, it may be useful to give the user some flexibility in specifying the query on images (i.e. the "importance" of different parts of the query). The possibility to specify imprecise queries in a system can significantly increase the effectiveness in the retrieval process on image databases.