Research And Development Of Feature-Based Multimedia Information Retrieval Technology

With the rapid development of the computer, network and communication technologies, the amount and types of information are booming day by day. How to get the required information from this huge multimedia library is becoming a very important subject, and more and more attentions have been paid by scholars around the world. According to this requirement, design and implementation of multimedia database are investigated in this thesis in detail. Combining the latest development achievements on database and multimedia technologies, the paper mainly studies the design and implementation of multimedia database in aspects of feature-based image retrieval and multimedia data model etc.In this paper, firstly multimedia database theory and method are discussed in detail and a summary is given about key technologies of the realization of multimedia database such as visit of database, disposal of digital images, identification of images,technologies of object-oriented and data model. At the same time the dissertation studies the application of these technologies. Secondly, the thesis researches the structure and retrieval procedure of feature-based retrieval in detail. Based on these studies, the paper puts forward a new retrieval method which improved the retrieval precision effectively with the integrated utilization of both color and texture characters. Finally the paper talks over the design and implementation of the Multimedia Database System.The system includes five modules: image query module, image feature extraction module, similitude match module, multimedia database management module and multimedia feature display module. C/S configuration mixed with B/S configuration is adopted in this system. When managing and maintaining the system within C/S configuration the user is also able to issue part of the multimedia resources on the internet by the B/S configuration.