Semantics Retrieval by Fuzzy Spatial Context of Image Objects
Hui Liu, Jun Ma, Jingsheng Lei, Ling Song · 2007
Modeling spatial context (e.g., autocorrelation) is a key challenge in classification and retrieval problems in image processing. In order to cope with the ambiguity of spatial relative position concepts, a new definition of the relative position between two objects in a fuzzy set framework is proposed according to traditional Markov random field (MRF). This definition is based on a fuzzy pattern-matching approach, and consists of comparing an object to a fuzzy set representing the degree of position satisfaction to a reference object. Fuzzy attributed relational graphs (FARGs) are used in this framework to evaluate the similarity between two images. Finally, we compare the image retrieval performance in several standard measures with MRF and FGM.