Interpretation of Fabric Tactile Perceptions through Visual Features for Textile Products

Abstract In this article, a systematic method based on two sensory experiments has been proposed to explore the relations between the visual and haptic perceptions of the tactile properties of a number of textile samples. The first experiment is designed to investigate to what extent fabric tactile properties could be perceived through the samples' visual representations. On this basis, the second experiment is carried out to study the mathematical relation between different sensory (visual and tactile) datasets. In this study, a new approach has been proposed based on rough inclusion degree and fuzzy inference system to deal with the sensory relations. The results show that for the discussed samples, most fabric tactile properties could be well perceived via vision. With the help of a fuzzy neural model, it is possible to predict a certain tactile property from the numerical scores of its principal visual features. Practical Applications Compared with researches focused on the development of mechanical simulative devices, the approach proposed here is much less expensive and could provide more direct and reasonable interpretation of fabric tactile information as both two sides of the model are from sensory systems which work as a whole during perception. Based on our model, with the help of image processing techniques, visual attributes will be automatically extracted from static or dynamic images, and interpreted by our rules to fabric tactile properties. Besides, this study will serve as a fundamental to the study on virtual fitting technologies in that the manufacturers or designers will be able to actively manipulate the virtual displays of textile products to provide our consumers with the most close‐to‐real sense of fabric.