Title: A new bag-of-words model using multi-cue integration for image retrieval

Authors: Junfeng Wu; Zhiyang Li; Wenyu Qu; Yuanyuan Li

Addresses: School of Information Science and Technology, Dalian Maritime University, Dalian, China; School of Information Engineering, Dalian Ocean University, Dalian, China ' School of Information Science and Technology, Dalian Maritime University, Dalian, China ' School of Computer Software, Tianjin University, Tianjin, China; School of Information Science and Technology, Dalian Maritime University, Dalian, China ' School of Information Science and Technology, Dalian Maritime University, Dalian, China; School of Software, Dalian Jiaotong University, Dalian, China

Abstract: Bag-of-words-based (BoW) image retrieval has been the most popular method in the image retrieval field. In traditional BoW scheme, the whole scheme for image retrieval utilises only one feature. The images are represented by the visual vocabulary consisting of a single feature. But in many cases, one feature could not describe the images well, some false matches often occur in traditional BoW scheme. To solve the problem, the paper proposes a multi-cue framework to integrate colour and shape features in order to improve the retrieval performance. To demonstrate the effectiveness of our method, we evaluate our new framework on the Simplicity dataset and Stanford dataset. The experimental results illustrate the proposed method could significantly improve the retrieval performance.

Keywords: bag-of-words; BoW; SIFT descriptor; saliency maps; colour histograms; multi-cue integration; image retrieval; colour features; shape features.

DOI: 10.1504/IJCSE.2016.077750

International Journal of Computational Science and Engineering, 2016 Vol.13 No.1, pp.80 - 86

Received: 27 Jul 2015
Accepted: 16 Oct 2015

Published online: 14 Jul 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article