Pictorial indexing for an integrated pictorial and textual IR environment

Information retrieval (IR) systems were originally developed to process, store, search and retrieve narrative information. Structured databases are therefore created to store the items containing enough information to identify and retrieve the onginal documents. This paper proposes an extension of classification and indexing to pictorial data, specifically to pictorial parts of multimedia documents. This goal is attained by adopting structural techniques for digital image description. The descriptions of the objects (digital structures) present in the image are proposed as pictorial index terms. Such a generalization has led to a uniform management of the non-homogeneous kinds of data composing the document and has allowed the outlining of a multimedia IR system raised from interdisciplinary experiences. In the paper the approach to image description is presented and a control structure to index textual and pictorial data is discussed.