An innovative microprocessor-based system for Human Activity Recognition: A fast and reliable classification algorithm

Introduction: The aim of this study is to describe the implementation of a microprocessor-based system for Human Activity Recognition (HAR) able to recognize seven activities of daily life (ADLs). The system is based on a body-worn miniature inertial measurement unit (IMU) that is part of a microprocessor system based on a 32-bit CPU (ARM 4 Cortex). A Decision Tree (DT) classifier implemented on the microcontroller performs the classification of the daily activities.

Source: Gait and Posture - October 1, 2018 Category: Orthopaedics Authors: G. De Leonardis, D. Fortunato, C.M. Gianfreda, S. Rosati, G. Balestra, M. Knaflitz Source Type: research