Cyclic Fusion of Measuring Information in Curved Elastomer Contact via Vision-Based Tactile Sensing

Abstract

Vision-based tactile sensors encode object data via optical signals, capturing microscale deformations using elastomer through densely arranged optical imaging sensors to detect subtle data variations. To enable continuous contact recognition, elastomers are crafted with curved surfaces to adjust to changes in the contact area. However, this design leads to uneven deformations, distorting tactile images and inaccurately reflecting the true elastomer deformations. In this work, we propose a cyclic fusion strategy for vision-based tactile sensing for precise contact data extraction and shape feature integration at the pixel level. Utilizing frequency-domain fusion, the system merges topography as indicated by elastomer deformation, enhancing information content by 8%, and regional information bias is reduced by 20% when preserving structural consistency. Furthermore, this system could effectively extract and summarize microscale contact features, decreasing erroneous predictions by 20% in defect detection via neural networks and reducing surface projection bias by 50% in surface depth reconstruction. Using this strategy, the measurement minimizes data interference, accurately depicting object morphology on tactile images and enhancing tactile sensation restoration.