Texture Selection Method using Category Classification

Abstract

For quantitative assessments of tactile qualities, many previous studies have used subjective ratings in terms of several pairs of adjective words (i.e., Semantic Differential method). However, this method requires observers to report their perceptual experiences one by one, and it takes substantial efforts, when the number of objects and/or the number of adjectives is increased. This research is, therefore, aimed at establishing a simple principle for analyzing the relationship of tactile qualities among objects, and selecting objects based on the differences of the qualities efficiently. Our principle asks observers only to classify objects into several groups according to the similarities of their tactile qualities. We can generate a distribution diagram and tree diagram of the objects, and then select appropriate objects on the basis of similarities of the tactile qualities.