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

摘要： 为提高语义图像分类器性能，提出一种基于公理化模糊集的语义图像层次关联规则分类器。首先，为提高算法精度，在对图像数据集进行特征提取基础上，采用公理化理论（AFS）构建图像集模糊概念的AFS属性表达，提高图像集属性辨识度；其次，为提高算法计算效率，考虑采用层次结构关联规则，构建语义图像分类器，利用概念之间的本体信息，提高并行分类能力；最后，通过对算法参数及横向对比实验，显示所提算法具有较高的计算精度和计算效率。

关键词: 公理化理论, 语义图像, 模糊集, 层次分类, 关联规则