Tuesday, September 29, 2009

Semantic Classification of Byzantine Icons

The painters of the Byzantine and post Byzantine artworks use specific rules and iconographic patterns for the creation of sacred figures. Based on these rules, the sacred figure depicted in the artwork is recognizable. In this work, we propose an automatic knowledge-based image analysis system used for Byzantine icons classification on the basis of the sacred figure recognition.

Figure 1. The Byzantine icon classification system’s architecture. The analysis module extracts information about the icon, and the knowledge representation and reasoning module uses this information to infer implicit knowledge that categorizes the icon.

Firstly, the system detects and analyzes the most important facial characteristics providing rich, still imprecise information about the Byzantine icon. Then, the information extracted is expressed in terms of an expressive terminology formalized using Description Logics (DLs) which form the basis of Semantic Web ontology languages.

Figure 2. Image analysis. First, the algorithm detects the sacred figure’s face region, eyes, and nose. Then, it extracts the hair, forehead, cheek, mustache, and beard parts together with the face’s base color layers. Further analysis of the extracted parts provides information about characteristic features. Finally, the algorithm produces a semantic interpretation for each of these features, together with formal assertions.

In order to effectively handle the imprecision involved, fuzzy extensions of DLs are used for the assertional part of the ontology. In this way, the extracted by image analysis information comprises the assertional component while the expressive terminology, formalizing the rules and the iconographic patterns, permits categorization of Byzantine artworks.

Figure 3. A reasoning example. The extracted information from image analysis constitutes the assertional component (ABox) of the knowledge base; the terminological component (TBox) is defined on the basis of Fourna’s specification. These components form the input to the fuzzy reasoning engine, which infers information about the icon.

P. Tzouveli, N. Simou, G. Stamou, S. Kollias. Semantic Classification of Byzantine Icons. In IEEE Intelligent Systems, vol 24, no. 2, pp. 35-43, March, April 2009. [ Abstract ] [ Bibtex ]

No comments: