A unified framework for semantics and feature based relevance feedback in image retrieval systems
MULTIMEDIA '00 Proceedings of the eighth ACM international conference on Multimedia
Content-Based Image Retrieval at the End of the Early Years
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
A personalized CBIR system based on a unified framework of fuzzy logic is proposed in this study. The user preference in image retrieval can be captured and stored in a personal profile. Thus, images that appeal to the user can be effectively retrieved. Our system provides users with textual descriptions, visual examples, and relevance feedbacks in a query. The query can be expressed as a query description language, which is characterized by the proposed syntactic rules and semantic rules. In our system, the semantic gap problem can be eliminated by the use of linguistic terms, which are represented as fuzzy membership functions. The syntactic rules refer to the way that linguistic terms are generated, whereas the semantic rules refer to the way that the membership function of each linguistic term is generated. The problem of human perception subjectivity can be eliminated by the proposed profile updating and feature re-weighting methods. Experimental results have proven the effectiveness of our system.