TPM
Textile Pressure Mapping Matrix
Concept
This project systematically investigates using planar pressure distribution sensing technology for ubiquitous and wearable activity recognition purposes. We propose a generic Textile Pressure Mapping (TPM) Framework, which encapsulates (1) design knowledge and guidelines, (2) a multi-layered tool including hardware, software and algorithms, and (3) an ensemble of empirical study examples. Through validation with various empirical studies, the unified TPM framework covers the full scope of application recognition, including the ambient, object, and wearable subspaces.
Hardware & Architecture
The framework enables TPM on both the mobile scale and large scale.
The mobile scale is implemented with microcontrollers with multichannel ADCs, with wireless communication, supporting up to 32x32 matrix @ 40fps.
The large scale is designed with FPGAs and high performance parallel ADCs, the largest implementation supports 128x128 matrix @ 40fps. Theoretically the architecture can be further scaled up.
Software Toolkit
A unified software toolkit which can be used across different design subspaces. The software implementation needs to consider the flexibility during algorithm exploration, as well as the real-time performance. The overall TPM software toolkit structure can be divided into 4 tiers: driver, pre-procesing, machine learning, output and feedback.
The software is implemented both in a heterogeneous environment with C++, Python, QML or HTML/CSS/JS, and an all-in-one executable in C++ utilizing openCV.
Algorithm & Feature Set
Includes generic algorithms that are shared across various empirical aspects and specific algorithms dealing with special application tasks.
The TPM feature set is designed to discriminate spatial-temporal data.
Transfer learning using deep neural networks is further evaluated for activity recognition using TPM data.