Summary
International Symposium on Nonlinear Theory and Its Applications
2015
Session Number:B4L-D
Session:
Number:B4L-D-2
Similarity-Based Image Retrieval by Self-Organizing Map with Refractoriness --- New Distance Between Image Features ---
Hideto Nakajima, Yuko Osana,
pp.708-711
Publication Date:2015/12/1
Online ISSN:2188-5079
DOI:10.34385/proc.47.B4L-D-2
PDF download (889.8KB)
Summary:
In most of the conventional image retrieval systems, the Euclidean distance is used in order to compare image features. However, in the Euclidean distance, 1 (which means the feature is included) and 0 (which means the feature is not included) are treated equally. In this paper, we propose a similarity-based image retrieval by self-organizing map with refractoriness using new distance between images features. We carried out a series of computer experiments and confirmed the effectiveness of the proposed system.