Smart Pressure e-Mat for Human Sleeping Posture and Dynamic Activity Recognition

TL;DR

A Smart Pressure e-Mat (SPeM) system based on piezoresistive material, Velostat, for human monitoring applications, including recognition of sleeping postures, sports, and yoga, demonstrating the high accuracy and generalizability of the models.

Abstract

With the emphasis on healthcare, early childhood education, and fitness, non-invasive measurement and recognition methods have received more attention. Pressure sensing has been extensively studied because of its advantages of simple structure, easy access, visualization application, and harmlessness. This paper introduces a Smart Pressure e-Mat (SPeM) system based on piezoresistive material, Velostat, for human monitoring applications, including recognition of sleeping postures, sports, and yoga. After a subsystem scans the e-mat readings and processes the signal, it generates a pressure image stream. Deep neural networks (DNNs) are used to fit and train the pressure image stream and recognize the corresponding human behavior. Four sleeping postures and 13 dynamic activities inspired by Nintendo Switch Ring Fit Adventure (RFA) are used as a preliminary validation of the proposed SPeM system. The SPeM system achieves high accuracies in both applications, demonstrating the high accuracy and generalizability of the models. Compared with other pressure sensor-based systems, SPeM possesses more flexible applications and commercial application prospects, with reliable, robust, and repeatable properties.

Figures (11)

• Figure 1: Schematic Illustration of the SPeM System for Human Activity Recognition. Different human activities generate distinct pressure distributions, which in turn result in varied resistance distributions. These resistance distributions subsequently give rise to diverse voltage distributions, enabling the generation of a stream of pressure images classified into different activities by a deep neural network.
• Figure 2: Design Diagram of Proposed SPeM System. The human pressure modality is represented as pressure distributions on the pressure e-mat. The signal processing subsystem generates a voltage distribution by scanning the interface of the pressure e-mat. The backend operator generates pressure images and utilizes a classifier for classification.
• Figure 3: Schematic Diagram of e-Mat. (a) Structure, size, and sensing area. (b) Size and spacing of sensor elements.
• Figure 4: Actual Image of e-Mat. (a) The e-mat is placed on a Queen size mattress. (b) The e-mat is folded and placed on a digital weight scale, and the total weight is 2749 grams.
• Figure 5: Fusion of ResNet-PIs for Images with Temporal Relationship. The DNN with three alternate architecture can be considered as 3D CNN, CRNN, and CDNN when concatenated with Conv, LSTM, or Dense, respectively.