A Content-Based Image Retrieval System Using Multiple Hierarchical Temporal Memory Classifiers

摘要：

This paper presents a content-based image retrieval (CBIR) system using multiple Hierarchical Temporal Memory classifiers. In order to improve the efficiency of image management and retrieval, the CBIR system proposed in this paper take advantage of Hierarchical Temporal Memory Algorithm which replicates the structure and function of the human neocortex. In this study, multiple Hierarchical Temporal Memory classifiers were used to provide an intelligent system that aims to understand a query images category semantics, rather than the low-level image features for image indexing and retrieval. The system supports query by example image, the experiments based on Internet images show the efficiency of our method.

关键词： Content-based image retrieval Hierarchical Temporal Memory Artificial Intelligence Image classification

作者: Xia Zhituo Ruan Hao Wang Hao

作者单位: Shanghai Institute of Optics and Fine Mechanics Chinese Academy of Sciences Shanghai, P.R. China

会议类型: 国际会议

会议名称: 2012 Fifth International Symposium on Computational Intelligence and Design 第五届计算智能与设计国际会议 ISCID 2012

会议地点: 杭州

会议语种:英文

页码: 1016-1019

在线出版日期: 2012-10-28（万方平台首次上网日期，不代表论文的发表时间）