Human Postures Recognition by Accelerometer Sensor and ML Architecture Integrated in Embedded Platforms: Benchmarking and Performance Evaluation

Abstract

Embedded hardware systems, such as wearable devices, are widely used for health status monitoring of ageing people to improve their well-being. In this context, it becomes increasingly important to develop portable, easy-to-use, compact, and energy-efficient hardware-software platforms, to enhance the level of usability and promote their deployment. With this purpose an automatic tri-axial accelerometer-based system for postural recognition has been developed, useful in detecting potential inappropriate behavioral habits for the elderly. Systems in the literature and on the market for this type of analysis mostly use personal computers with high computing resources, which are not easily portable and have high power consumption. To overcome these limitations, a real-time posture recognition Machine Learning algorithm was developed and optimized that could perform highly on platforms with low computational capacity and power consumption. The software was integrated and tested on two low-cost embedded platform (Raspberry Pi 4 and Odroid N2+). The experimentation stage was performed on various Machine Learning pre-trained classifiers using data of seven elderly users. The preliminary results showed an activity classification accuracy of about 98% for the four analyzed postures (Standing, Sitting, Bending, and Lying down), with similar accuracy and a computational load as the state-of-the-art classifiers running on personal computers.

The Effect of using Dimensionality Reduction Compared with Type of Algorithm on Detecting Patient Fall: Triage Case Study

“…In [27] the authors developed an automatic system based on a three-axis accelerometer was developed to recognize the four analyzed postures (standing, sitting, bending, and lying down), which is useful in detecting potentially inappropriate behavioral habits in the elderly. The system used ML algorithms to recognize postures in real-time and has been optimized to perform highly on low-computational power and energy-consuming platforms.…”

Exploring the Use of Contrastive Language-Image Pre-Training for Human Posture Classification: Insights from Yoga Pose Analysis

“…They are generally small, accurate, and robust to external conditions. Due to these features, they are still used for some specific applications and are preferred to cameras [7,8]. However, they also have some drawbacks, such as being more expensive, more cumbersome, more prone to noise and drift, and less scalable to multiple individuals.…”

IoT System for Real-Time Posture Asymmetry Detection

“…For example, in the context of smart homes and assisted living, Stern et al [ 17 ] recently described a system for in-bed posture recognition by convolutional neural networks based on the pressure data acquired from a pressure mat positioned on the mattress aimed at investigating the relation between the sleeping positions and the insurgence of pressure sores in bedridden people. Leone et al [ 18 ] developed a posture recognition machine learning algorithm for the classification of four postures: Standing, sitting, bending, and laying down. Concerning the monitoring of workforce health conditions, Lind et al [ 19 ] overviewed the available systems based on motion capture instruments to collect kinematics data for the prevention of work-related musculoskeletal disorders.…”