Ontology-Based Image Information Organization And Use In Network Learning Resource

With the development of computer technology, internet, and multi-media, as well as the mention of semantic web technology, network has become an important platform for us to learn. The sharp growth of multimedia data, especially digital cameras and mobile phone photograph popularization, the number of image information is growing rapidly. Image can express information vividly and directly, and can leave us a deep impression, so it has become an indispensable part of online learning resources.In recent years, in information retrieval technology field, theory of ontology causes more and more extensive research. This thesis firstly introduces network learning resources, semantic web and ontology knowledge, analyzes the relevant research in the application of ontology in image retrieval field and then constructs a ontology model which is generic, able to fully describe image and could be used to describe image's all aspects of semantic features. In addition , this thesis introduces the basic theory of support vector machine, mainly to achieve automatic extraction of image's high-level semantic features, and using support vector machine is the bottom of the image of features mapping to ontology, thus realize high-level semantic concept from the bottom of image characteristics of the automatic acquisition image semantic comments information. Finally, this thesis realize an image management experiment system and image retrieval experiment system based on ontology, and using the system to verify the effect of two retrieval way. The experimental results show that the proposed image retrieval algorithm greatly improves the precision and recall.