论文部分内容阅读
首先对当前数字图书馆中多媒体信息检索的现状进行分析,然后讨论了跨媒体检索的优点及应用前景。挑选以澳大利亚国家图书馆“Trove系统”为代表的传统跨媒体检索模型和浙江大学开发的“百万册图书馆”项目为代表的改进的跨媒体检索模型分别进行深入分析。重点分析这两大模型的实现原理、技术方案。最后,对比这两大模型的优缺点,并从3个角度就模型的改进和优化提出了构思及实现方案,并对这些方案进行定性分析。
First of all, the current situation of multimedia information retrieval in digital library is analyzed, and then the advantages and application prospect of cross-media retrieval are discussed. The paper chooses the improved cross-media retrieval model represented by the traditional cross-media retrieval model represented by “Trove System” of Australian National Library and the “Million Books” project developed by Zhejiang University. Focus on the analysis of the realization of these two models theory, technical solutions. Finally, the advantages and disadvantages of these two models are compared, and the conception and implementation plan of the model improvement and optimization are proposed from three aspects, and the qualitative analysis of these plans is made.