Adaptive Image Retrieval Based on Multi-Feature Fusion

DOI

10.2991/cnci-19.2019.84How to use a DOI?

Keywords

Image retrieval, adaptive, multi-feature fusion, information entropy.

Abstract

With the rapid development of technology and the popularity of the Internet, a large number of image datasets have been produced in various industries. It is difficult to quickly search for a desired image in a large number of image datasets, so it is very meaningful to design an efficient image retrieval system. This paper proposes an adaptive weighting method based on information entropy. Firstly, the trust of a single feature is obtained according to the information entropy. Then the transfer matrix is constructed according to the trust of a single feature, and the weight of the single feature is obtained by iterative calculation according to the transfer matrix. This method has the following advantages: (1)The retrieval system combines the performance of multiple features and has higher accuracy than the retrieval of a single feature;(2) In the query process, the weight of the individual features is dynamically updated using the query image, so that the retrieval system takes full advantage of the single feature. The experimental result show that the proposed method is better than the previous method. The Mean Average Precision of the search on the Holidays image dataset is 83.99%.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Xiaoqian You
AU  - Jianghua Si
PY  - 2019/05
DA  - 2019/05
TI  - Adaptive Image Retrieval Based on Multi-Feature Fusion
BT  - Proceedings of the 2019 International Conference on Computer, Network, Communication and Information Systems (CNCI 2019)
PB  - Atlantis Press
SP  - 608
EP  - 612
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnci-19.2019.84
DO  - 10.2991/cnci-19.2019.84
ID  - You2019/05
ER  -