Automatic image annotation using ontology

Semantic annotation of images is choosing appropriate labels for images with extracting effective and hidden feature in pictures, in the proposed method features has extracted from image after segmentation. In addition to the inherent and conceptual features, spatial features of each segments and relation between them are also considered.Then these features with fuzzy rules are trained and in this train we use hierarchical structure for describing and labeling the images that in each of the levels in hierarchical structure used from SVM classification. Results on the images of Imageclef database (this database is classified with hierarchical structure) show the proposed method of image annotation in addition to reducing the complexity of the classification increased accuracy compared to the another methods by approximately 5%.