文献类型：期刊文献

英文题名：Region based image retrieval method

作者：Wang, Yao-Wei[1]; Zou, Li-Kun[2]; Su, Li-Min[3]; Luo, Sen-Lin[1]

第一作者：Wang, Yao-Wei

通讯作者：Wang, Y.-W.

机构：[1] Department of Electronic Engineering, School of Information Science and Technology, Beijing Institute of Technology, Beijing 100081, China; [2] Central Institute for Correctional Police, Baoding 071000, China; [3] School of Information, Beijing Union University, Beijing 100100, China

第一机构：Department of Electronic Engineering, School of Information Science and Technology, Beijing Institute of Technology, Beijing 100081, China

通讯机构：[1]Department of Electronic Engineering, School of Information Science and Technology, Beijing Institute of Technology, Beijing 100081, China

年份：2007

卷号：27

期号：SUPPL. 1

起止页码：97-99

外文期刊名：Beijing Ligong Daxue Xuebao/Transaction of Beijing Institute of Technology

收录：EI(收录号：20081611206576);Scopus(收录号：2-s2.0-41949142932)

语种：英文

外文关键词：Closed circuit television systems - Color - Robustness (control systems)

摘要：A region based image retrieval method is proposed. Not only local and overall features are used in the method, but spatial constraints arc also employed. Experimental data are composed of 1 000 images from Corel and 1000 I frames extracted from CCTV and MPEG-7 videos. Three methods namely the global color feature, color and spatial constraints, as well as the color/texture with constraints are compared. An amount of noise is introduced to test the robustness of the method. Experimental results show that the method is robust and effective.