Title: Voting algorithm into CBIR: a review

Authors: Mawloud Mosbah

Addresses: Informatics Department, Faculty of Science, LRES Laboratory, University 20 Août 1955 of Skikda, Algeria

Abstract: Collectiveness is becoming nowadays the trend and the ubiquitous notion available in almost recent introduced informatics smart systems like: collective intelligence tied to bio-inspired solutions and reactive multi-agent systems. In the context of collectiveness and collective decision-making, voting algorithm is an important alternative helping to achieve better accuracy and performance. As any other scientific field, CBIR has taken profit of many recently introduced techniques and methods including voting algorithm. In this paper, we talk about voting algorithm into CBIR through quoting its different utilisation with various operations and sub-operations over the different CBIR components such as ranking, re-ranking, feature extraction and representation, feature selection, clustering and classification. Although that the algorithm is the same in its processing mode, the different utilisation contexts lead to different versions especially regarding the considered resources and targets. The paper, presented as a review, allows then collecting and organising the different works adopting voting algorithm, with its different versions, into CBIR field, and discussing the performance and the efficiency of the algorithm in the different considered contexts.

Keywords: content-based image retrieval; CBIR; CBIR performance enhancement; CBIR pre-processing; CBIR post-processing; voting algorithm.

DOI: 10.1504/IJIM.2023.132311

International Journal of Image Mining, 2023 Vol.4 No.2, pp.124 - 145

Received: 04 Dec 2020
Accepted: 30 Jan 2022

Published online: 18 Jul 2023 *

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