Tactile modeling and rendering image-textures based on electrovibration

Abstract

Image-textures contain most surface features of real objects that are largely missing from virtual tactile interactions. This paper presents a tactile model for rendering image-textures based on electrovibration, which is achieved by varying stimuli signals to modulate friction between the finger and the touchscreen. We do research on the relationships between human tactile sensation and stimuli signals through experiments. According to the relationships, we establish a mapping model based on gradients of image-textures which are gained by the Roberts filter. We use the mapping model to synthetize the frequency and amplitude of stimuli signals for rendering image-textures as a user interacts with our tactile prototype. Specifically, stimuli frequency mainly reflects hardness and granularity of image-textures, stimuli amplitude mainly reflects heights of image-textures. We compare the proposed model with the model based on stimuli amplitude through experiments on the prototype. Results show that the proposed model can effectively enhance tactile reality of image-textures.