基于复杂网络的图像特征提取及多特征融合方 案探究.pdf

Journal of Image and Signal Processing  图像与信号处理, 2015, 4(4), 101‐110  Published Online October 2015 in Hans. /journal/jisp  /10.12677/jisp.2015.44012    Image Feature Extraction Based on Complex Network and Multi‐Feature Fusion Schemes Exploration in CBIR 1,2 1* 1 Jibin Gao , Yumei Li , Huina Zhang 1 Department of Mathematics, School of Science, Beijing Technology and Business University, Beijing  2 College of Science, Guilin University of Technology, Guilin Guangxi  * Email:  liwjyumei@163.com    rd th st Received: Oct. 3 , 2015; accepted: Oct. 16 , 2015; published: Oct. 21 , 2015  Copyright © 2015 by authors and Hans Publishers Inc.  This work is licensed under the Creative Commons Attribution International License (CC BY).  /licenses/by/4.0/    Abstract Image shape feature’s extraction is an important research content in content‐based image re‐ trieval, and an image shape feature extraction method by using complex network model is pro‐ posed in this paper. First, SIFT keypoints of an image are extracted, and then the image is divided into blocks such that the initial complex network model can be built in each block respectively. After that, minimum spanning tree decomposition method is used for the network’s dynamic evo‐ lution, and the network features at different moments in different blocks are extracted as the im‐ age’s shape features. Furthermore, the shape features are combined with the color and texture features and a kind of fusion feature is obtained. By experiment results comparison, it shows that