Using Fuzzy DLs to Enhance Semantic Image Analysis

Abstract

Abstract. Research in image analysis has reached a point where detectors can be learned in a generic fashion for a significant number of conceptual entities. The obtained performance however exhibits versatile behaviour, reflecting implications over the training set selection, similarities in visual manifestations of distinct conceptual entities, and appearance variations of the conceptual entities. In this paper, we investigate the use of formal semantics in order to benefit from the logical associations between the conceptual entities, and thereby alleviate part of the challenges involved in extracting semantic descriptions. More specifically, a fuzzy DL based reasoning framework is proposed for the extraction of enhanced image descriptions based on an initial set of graded annotations, generated through generic image analysis techniques. Under the proposed reasoning framework, the initial descriptions are integrated and further enriched at a semantic level, while additionally inconsistencies emanating from conflicting descriptions are resolved. Experimentation in the domain of outdoor images has shown very promising results, demonstrating the added value in terms of accuracy and completeness of the resulting content descriptions.

An overview of fuzzy Description Logics for the Semantic Web

“…However, in the Semantic Web, there is much imprecise and uncertain knowledge, which cannot be represented by these rule languages. To this end, based on OWL DL, that is, the DL SHOIN(D), Pan et al (2006a) Simou et al (2008aSimou et al ( , 2008b, Straccia and Visco (2007), Straccia (2010), Singh et al (2004) and Stoilos et al (2005d) Sanchez andYamanoi (2006), Stoilos et al (2005cStoilos et al ( , 2010Stoilos et al ( , 2006c, Gao and Liu (2005), Calegari and Ciucci (2007) and Bobillo and Straccia (2009c) f-SWRL: Pan et al Straccia (2006cStraccia ( , 2004b, Venetis et al (2007) and Zhao and Boley (2008) Lukasiewicz (2006) and Straccia (2007a, 2007b) Medicine: Molitor and Tresp (2000), D'Aquin et al (2006) and Schlobach et al (2007) Ontology mpping : Nova´cˇek and Smrzˇ(2006), Ferrara et al (2008) and Xu et al (2005) Electronic market: Ragone et al (2008aRagone et al ( , 2008bRagone et al ( , 2008cRagone et al ( , 2007 and Agarwal and Lamparter (2005) Data modeling: Zhang et al (2008Zhang et al ( , 2009) Semantic Web portals: Zhang et al (2005) Semantic search engines: Li et al (2007Li et al ( , 2006d Image analysis: …”

A Fuzzy Extension to the OWL 2 RL Ontology Language

“…. , B n }, called body, can be a set of atoms or the special atom t. 5 Moreover, each variable in H also occurs in the body, and all variables are implicitly assumed to be universally quantified. The semantics of fuzzy Datalog can also be given via fuzzy interpretations.…”

“…For example, in our hypothetical scenario one cannot capture the fact that doctor "a" is better than doctor "b" who is still quite good though. Managing fuzzy knowledge is of great importance in many applications, like multimedia processing [5,6], decision making [7], negotiation [8], and more. For these reasons many fuzzy extensions to OWL and DLs have been proposed [9,10,11,12,13,14,15,16,17].…”

Using Association Rule Mining to Enrich Semantic Concepts for Video Retrieval

“…In recent work of (Dasiopoulou et al, 2008), a reasoning framework based on fuzzy description logics is used to enhance the extraction of image semantics. Explicit semantic relationships among concepts are represented using assertions of description logics.…”