Image Retrieval System Using Japanese Sound Symbolic Words for Surface Texture

Koichi YAMAGATA; Tatsuki KAGITANI; Maki SAKAMOTO

doi:10.5057/isase.2019-C000027

ISASE2019

Session ID : 2-B-3

DOI https://doi.org/10.5057/isase.2019-C000027

Conference information

Host: Japan Society of Kansei Engineering

Name : The 5th International Symposium on Affective Science and Engineering

Number : 5

Location : Kogakuin University

Date : March 17, 2019 - March 18, 2019

2-B: Artificial Robotics and Computing

Image Retrieval System Using Japanese Sound Symbolic Words for Surface Texture

Koichi YAMAGATA, Tatsuki KAGITANI, Maki SAKAMOTO

Author information

Keywords: onomatopoeia, sound symbolism, texture impression

CONFERENCE PROCEEDINGS FREE ACCESS

Details

Abstract

We propose a system to retrieve appropriate images from a database for a given Japanese sound symbolic word (SSW) expressing texture. We use some features of images calculated from gray-level co-occurrence matrices (GLCM). Using GLCMs and SSWs contribute reduction of calculation cost and easy operation of the system.

Corresponding author

Register with J-STAGE for free!