关键词: 公理化理论, 语义图像, 模糊集, 层次分类, 关联规则

Abstract: In order to improve the performance of semantic image classification, the semantic image classifier based on hierarchical association rule with axiomatic fuzzy set is proposed. Firstly, in order to improve the accuracy of the algorithm, the image data set for feature extraction is constructed based on the axiomatic theory（AFS） to realize AFS image sets fuzzy concept expression, which improves the image set attribute recognition. Secondly, in order to improve the computational efficiency of the algorithm, the hierarchical structure association rules are considered, and it constructs the semantic image classifier, which uses the ontology information to improve the ability of parallel classification. Finally, through the comparison of the algorithm parameters and the horizontal contrast, the results show that the proposed algorithm has high accuracy and computational efficiency.

Key words: axiomatic theory, semantic image, fuzzy set, hierarchical classification, association rule