文献类型：期刊文献

英文题名：Representation of image content based on Rol-BoW

作者：Zhang, Jing[1,2];Li, Da[1];Zhao, Yaxin[1];Chen, Zhihua[1];Yuan, Yubo[1]

机构：[1]E China Univ Sci & Technol, Dept Comp Sci & Engn, Shanghai 200237, Peoples R China;[2]Nanjing Univ, State Key Lab Novel Software Technol, Nanjing 210093, Jiangsu, Peoples R China

年份：2015

卷号：26

起止页码：37

外文期刊名：JOURNAL OF VISUAL COMMUNICATION AND IMAGE REPRESENTATION

收录：;WOS:【SCI-EXPANDED(收录号:WOS:000348249000005)】；

基金：This research has been supported by the National Nature Science Foundation of China (Grant 61402174 and 61370174), and Nature Science Foundation of Shanghai Province of China (11ZR1409600).

语种：英文

外文关键词：Rol-BoW; Representation of image content; Different size segmentation; Image retrieval; Bag of words; Region of Interest; Feature extraction; Gabor filtering

摘要：Representation of image content is an important part of image annotation and retrieval, and it has become a hot issue in computer vision. As an efficient and accurate image content representation model, bag-of-words (BOW) has attracted more attention in recent years. After segmentation, BoW treats all of the image regions equally. In fact, some regions of image are more important than others in image retrieval, such as salient object or region of interest. In this paper, a novel region of interest based bag-of-words model (Rol-BoW) for image representation is proposed. At first, the difference of Gaussian (DOG) is adopted to find key points in an image and generates different size grid as Rol to construct visual words by the BoW model. Furthermore, we analyze the influence of different size segmentation on image content representation by content based image retrieval. Experiments on Corel 5K verify the effectiveness of Rol-BoW on image content representation, and prove that Rol-BoW outperforms the BoW model significantly. Moreover, amounts of experiments illustrate the influence of different size segmentation on image representation based on the Bow model and Rol-BoW model respectively. This work is helpful to choose appropriate grid size in different situations when representing image content, and meaningful to image classification and retrieval. (C) 2014 Elsevier Inc. All rights reserved.