An Analysis on Color Characteristics of Website Images of Restaurants According to Price Range

Abstract

In this study, we propose a method of Kansei analysis of aspects such as preference or impression using a large amount of automatically processed data. We focused on representative colors which are low colors representing many colors in an image because colors affect to people’s feeling and impression and are easily quantify and express features. Our method consisted of automatic data acquisition by web crawling and automatic feature extraction through an algorithm of image processing. Data collected from restaurant websites were analyzed in order to verify the following hypothesis: expensive restaurants have more achromatic photographs on their websites. As a result, the hypothesis has been proven in several restaurant genres.