Title :
Application of Fourier spectrum for describing ´formal´ of cloths
Author :
Asano, Chie Muraki ; Muneyasu, Mitsuji ; Ozeki, Takashi ; Asano, Akira ; Fujimoto, Takafumi
Author_Institution :
Hiroshima Univ., Japan
Abstract :
Summary form only given. Quantitative evaluation of human sensibility, that is, "kansei" in Japanese, has recently attracted much attention. Kansei engineering is an approach to connect the human sensibility with engineering applications. We have investigated a relationship between the visual impressions of black fabrics and their image features from the viewpoint of kansei engineering. The two-dimensional Fourier transformation was applied for analyzing an image of black fabric in order to connect with formality, which is one of evaluation items for the visual impression of fabrics. A sensory vision test was carried out on several typical kinds of black fabrics, while Fourier spectrum features were extracted from the fabric images. Amplitudes of the Fourier spectrum were classified by the k-means method, which is a cluster analysis method, and were discussed to analyze the visual impression. This study has found that differences of formality in the visual impressions of texture characteristics are derived from the features of the classified amplitudes of the Fourier spectrum.
Keywords :
Fourier transforms; fabrics; feature extraction; image classification; image texture; pattern clustering; visual perception; 2D Fourier transform; Fourier spectrum amplitude classification; Fourier spectrum feature extraction; black fabrics; cluster analysis method; fabric formality; human sensibility; image features; k-means method; kansei engineering; sensory vision test; texture characteristics; two-dimensional Fourier transform; visual impressions; Amplitude estimation; Computer simulation; Fabrics; Frequency estimation; Humans; Image processing; Knowledge based systems; Nonlinear filters; Pediatrics; Speech analysis;
Conference_Titel :
Nonlinear Signal and Image Processing, 2005. NSIP 2005. Abstracts. IEEE-Eurasip
Conference_Location :
Sapporo
Print_ISBN :
0-7803-9064-4
DOI :
10.1109/NSIP.2005.1502216