Designing a new deep convolutional neural network for content-based image retrieval with relevance feedback. (March 2023)

• Record Type: Journal Article
• Authors: Rastegar, Homayoun; Giveki, Davar
• Is Part Of: Computers & electrical engineering. Volume 106(2023)
• Publication Date: 2023-03
• Subjects: Content-based image retrieval -- Relevance feedback -- Deep learning -- Convolutional neural network -- Generalized discriminant analysis; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2023.108593
• Languages: English
• ISSNs: 0045-7906

Abstract

Abstract: Content-based image retrieval (CBIR) is a hot research topic in computer vision. Relevance Feedback (RF) is a powerful technique that can help to increase quality of the CBIR. In this study, a new method for addressing the problem of the CBIR is proposed. To this end, a novel Convolutional Neural Network (CNN) and a new framework of applying RF have been investigated. Also, in order to obtain more efficient image features and reduce the dimensionality of feature vectors, we applied Generalized Discriminant Analysis (GDA) on the extracted features. The proposed method was tested on three benchmark datasets including the OT, Corel-1000 and Caltech-101. The experimental results illustrate that the proposed method performs better compared to the state-of-the-art in the field. Our approach achieved remarkable performances in image retrieval by demonstrating mean Average Precisions (mAPs) of 96.8%, 95.62% and 98.78% on the OT, Corel-1000 and Caltech-101, respectively.