Content Based Image Retrieval: Contemporary Trends and Challenges

Bajrami, Buen and Kotevski, Zoran and Veljanovska, Kostandina (2023) Content Based Image Retrieval: Contemporary Trends and Challenges. In: International conference on Applied Internet and Information Technologies - AIIT2023, 13th October 2023, Bitola, North Macedonia.

This is the latest version of this item.

[thumbnail of Paper_AIIT2023.pdf] Text

Download (3MB)


Content Based Image Retrieval (CBIR) is a process that enables finding similar images in large sets of databases based on an image query. The purpose of CBIR engines is to mimic humans in the image classification process, thus a high computational cost is required to reach the proper selection of features which must be as unique as possible. Otherwise, the wrong choice of the model and then of the image features negatively affects the results by providing images that are not so similar to query image. In the last two decades, the number of researches related to CBIR has increased, but even today it is considered as difficult process. For this reason, the researchers in this field have developed various models and techniques that help in rendering the image, trying to make the results as accurate as possible. Some of these techniques include Local Binary Pattern (LBP) histogram, Local Difference Binary (LDB), Local Tetra Pattern (LTrP), etc. In this paper, we review the latest research in the field of CBIR, we compare their performance based on several factors, such as calculation time, image acquisition time, accuracy of results and we conclude the paper with a discussion about which models have shown the best performance and what are their advantages and disadvantages for which we make some recommendations for the future research.

Item Type: Conference or Workshop Item (Paper)
Subjects: Scientific Fields (Frascati) > Natural sciences > Computer and information sciences
Scientific Fields (Frascati) > Engineering and Technology > Electrical engineering, electronic engineering,information engineering
Divisions: Faculty of Information and Communication Technologies
Depositing User: Prof. d-r. Zoran Kotevski
Date Deposited: 10 Dec 2023 20:23
Last Modified: 10 Dec 2023 20:23

Available Versions of this Item

Actions (login required)

View Item View Item