结合低层特征和高层语义的图像检索系统-模式识别与智能控制专业论文.docxVIP

西北工业大学硕士论文摘要 西北工业大学硕士论文 摘要 计算机技术、多媒体技术以及Intemet技术的飞速发展产生大量的图像信息， 因此如何有效地、快速地从大规模的图像数据库中检索出需要的图像是目前一个 急需解决的重要问题。基于内容的图像检索技术和基于语义的图像检索技术正是 解决这一问题的有效途径。前者研究的是根据自动荻取的图像低层特征，从图像 数据库中检索出相关图像；而后者研究的是如何从多种渠道获取图像语义信息， 并根据语义检索相关图像。 本论文主要围绕图像低层特征的提取和索引、高层语义特征的检索、弥补语 义鸿沟的相关反馈技术展开详细的研究。并在研究的基础上，结合三项技术的优 点，建立了高效实用的图像检索系统。此系统为图像构建语义关键字网络，并建 立低层特征库，对图像实行特征和语义的复合索引；通过相关反馈技术一方面在 线理解用户的查询意图，自动调整相似度测量准则以符合用户需求；另一方面给 相关图像传递语义标注，更新相关性强度，充实语义网络，实现长期的学习和记 忆。在检索的过程中，用户可以通过多种不同途径给出查询条件，图像的低层特 征与语义网络互为补充以达到最佳的检索效果。经过试验测试，本丈的图像检索 系统可以在少量的反馈次数(平均5次)中达到进70％的准确率，而且随着系统 的使用次数的增多，检索性能会逐步增长。 关键字基于内容的图像检索，语义图像检索，相关反馈， 关键字网络 西北工业太掌硕士论文ABSTRACT 西北工业太掌硕士论文 ABSTRACT The rapid development of computer,multimedia and Intemet techniques has produced too large amount of images．Therefore，it becomes all urgent problem that how to find needed image efficiently in large-scale image database．Two effective ways has been proposed to solve the problem：one is content—based image retrieval technique which search target images by low-level content feature．The other iS semantic image retrieval technique which in search of ways to get images’semantic infotination． In this paper,a lot research work has been done around three points：how to abstract and index low-level feature of images，how to retrieval images in semantic level，and how to fill the semantic gap by relevant feedback technique．Based on the research work，An efficient and practical image retrieval system is built which integrated the advantages of these techniques． In the system．key WOrd networks and lOW—level feature table are both established for images，which realized double indexing；relevant feedback is also applied in the system．On one hand，it enables the system catch users’query intention Oil Iine by adjusting its similarity criterion automatically；on the other hand，it pass the annotations for relevant images，update weights between key words and images and filI the semantic networks．thus tlle system can study in long．term．Users Can provide 廿Ieir query requir