Image Retrieval System Based On Multi-Feature Combination With The User Feedback

Along with the rapid development of computer network technology in today’s society, A large number of image information appears in people’s lives. Image data as the most common basic multimedia information, leading to the birth of many large-scale image database, and how to scientifically manage these image databases is concerned by more and more people, so the image retrieval has become a hot research field of the computer. Content-based image retrieval technology (Content-based Image Retrieval CBIR) uses Image color, texture and shape features for image database indexing, which reduces labor of manual annotation, even in different countries and between different peoples it can also establish uniform retrieval method. CBIR becomes a very active topic of information retrieval.In this paper, a generic content-based image retrieval system is implemented, this system extracts the image’s color, texture and shape features for image database indexing. And use a combination of the features to retrieval and can also optimize the user feedback. The paper includes:1. Research the information technology used by Content-based image retrieval system, provide a platform and some development tools used for the system and give a brief introduction of them. Finally the paper gives the other techniques and algorithms used by the system.2. Through in-depth analysis of the basic structure of the content-based image retrieval system, and through systematic analysis on this basis, establish the basic structure of the system. Then combined with the appropriate IT build the system model.3. The implementation of the system consists of two parts, Firstly, the basic realization of the image storage module, Secondly, the basic realization of the image retrieval module, through the process of realization deeply study the content-based image retrieval system. And summarizes the implementation of the system and the optimization direction.