| 基于流形学习的图像检索研究进展 |
| |
| 引用本文： | 刘利,韦佳,马千里.基于流形学习的图像检索研究进展[J].北方交通大学学报,2010(5):164-171. |
| |
| 作者姓名： | 刘利 韦佳 马千里 |
| |
| 作者单位： | [1]惠州学院计算机科学系,广东惠州516007 [2]华南理工大学计算机科学与工程学院,广东广州510641 |
| |
| 基金项目： | 广东省自然科学基金资助项目（07006474,94510641,01003233）;广东省科技攻关项目资助（2007B010200044）;华南理工大学中央高校基本科研业务费专项资金资助（2009ZM0189） |
| |
| 摘 要： | 基于内容的图像检索是当前研究的热点，然而由于“语义鸿沟”问题而限制其检索能力的提高．而流形学习可以利用图像数据库中的数据，以及和用户交互的反馈信息获得用户的语义概念，从而提高检索性能．本文鉴于流形学习在图像检索中表现出的有效性，分析了近几年将流形学习应用到基于内容图像检索中的算法，从归纳和转导角度将相关内容分为两个类别，针对每个类别总结分析了相关算法，并总结了有待进一步研究的问题．
|
| 关 键 词： | 计算机应用技术 流形学习 流形排序 基于内容的图像检索 语义子空间学习 |
State-of-the-Art on Image Retrieval Based on Manifold Learning |
| |
| LIU Li,WEI Jia,MA Qianli.State-of-the-Art on Image Retrieval Based on Manifold Learning[J].Journal of Northern Jiaotong University,2010(5):164-171. |
| |
| Authors: | LIU Li WEI Jia MA Qianli |
| |
| Institution: | 1. Department of Computer Science, Huizhou University, Huizhou Guandong 516007, China; 2. School of Computer Science and Engineering South, China University of Technology, Guangzhou Gnandong 510641, China) |
| |
| Abstract: | At present, content-based image retrieval is one of the hottest research fields. However, semantic gap has heavily limited its performance improvement. Manifold can use data sample in image database and relevance feedback from user to learn user＇s semantic conception to improve retrieval performance. In view of the effectiveness of manifold learning for the performance of image retrieval, we overviewed the existing methods that apply manifold learning to content-based image retrieval in recent years, and classified them from transductive and induction views. Finally, we discuss key issues that reel to be solved. |
| |
| Keywords: | computer application technology manifold learning manifold ranking content-based image retrieval semantic subspace learning |
| 本文献已被 维普 等数据库收录！ |
