Title: Parallel big image data retrieval by conceptualised clustering and un-conceptualised clustering

Authors: Ja-Hwung Su; Chu-Yu Chin; Jyun-Yu Li; Vincent S. Tseng

Addresses: Department of Information Management, Cheng Shiu University, No. 840, Chengcing Rd., Niaosong Dist., Kaohsiung, Taiwan ' Department of Computer Science and Information Engineering, National Cheng Kung University, No. 1, University Road, Tainan, Taiwan; Telecommunication Laboratories, Chunghwa Telecom Co., Ltd., No. 99, Dianyan Rd., Yangmei District, Taoyuan, Taiwan ' Department of Information Management, Cheng Shiu University, No. 840, Chengcing Rd., Niaosong Dist., Kaohsiung, Taiwan ' Department of Computer Science, National Chiao Tung University, 1001 University Road, Hsinchu, Taiwan

Abstract: Content-based image retrieval is a hot topic which has been studied for few decades. Although there have been a number of recent studies proposed on this topic, it is still hard to achieve a high retrieval performance for big image data. To aim at this issue, in this paper, we propose a parallel content-based image retrieval method that efficiently retrieves the relevant images by un-conceptualised clustering and conceptualised clustering. For un-conceptualised clustering, the un-conceptualised image data is automatically divided into a number of sets, while the conceptualised image data is divided into multiple sets by conceptualised clustering. Based on the clustering index, the depth-first-search strategy is performed to retrieve the relevant images by parallel comparisons. Through experimental evaluations on a large image dataset, the proposed approach is shown to improve the performance of content-based image retrieval substantially in terms of efficiency.

Keywords: content-based image retrieval; CBIR; un-conceptualised clustering; conceptualised clustering; big data; parallel computation.

DOI: 10.1504/IJHPCN.2019.103538

International Journal of High Performance Computing and Networking, 2019 Vol.15 No.1/2, pp.22 - 30

Received: 31 Mar 2018
Accepted: 19 Jul 2018

Published online: 11 Nov 2019 *

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