文献类型：期刊文献

英文题名：Image retrieval algorithm based upon co-occurrence matrix of color and edge orientation

作者：Zhang, Qing[1]; Lin, Jiajun[2]; Liu, Yunxiang[1]

机构：[1] Computer Science and Information Engineering Department, Shanghai Institute of Technology, Shanghai 201418, China; [2] Research Institute of Automation, East China University of Science and Technology, Shanghai 200237, China

年份：2013

卷号：9

期号：1

起止页码：171

外文期刊名：Journal of Computational Information Systems

收录：EI(收录号：20130515954100)

语种：英文

外文关键词：Behavioral research - Edge detection - Color - Image enhancement

摘要：In order to extract effectively features to improve the performance of image retrieval, a novel contentbased image retrieval algorithm based upon the color and edge orientation is proposed. It computes the co-occurrence matrices of intensity, four broadly-tuned color channels of Itti's visual attention model and the edge orientation, and then utilizes energy, contrast, entropy and homogeneity to describe the image features. The experiments show that our proposed algorithm distinguishes color, texture and shape in a better way and achieves high precision and recall rate. ? 2013 Binary Information Press.