A Structure Modality Enhanced Multimodal Imaging Method for Electrical Impedance Tomography Pressure Distribution Measurement

Abstract

Electrical impedance tomography (EIT) based pressure distribution sensors have the advantages of a simple structure and the ability to continuously measure pressure over a large area, making it a promising solution for large-scale artificial robotic skin. However, achieving high spatial resolution reconstruction of pressure distribution with EIT pressure sensors is challenging because the positions, sizes, and magnitudes of the pressure of the compressed areas are deeply coupled and mutually influenced in the EIT reconstructed results. To address this issue, a novel multimodal EIT pressure distribution measurement method is proposed. In this method, a structure modality EIT pressure sensor is designed to provide independent position and size information of the compressed areas to complement the pressure distribution measured using a normal EIT pressure sensor. A multimodal convolutional neural network (CNN) was designed to fuse the multimodal EIT sensors. The simulations and experiments demonstrate that the proposed multimodal EIT sensor outperforms the regular single-modality EIT sensor.