Human Daily Activity Recognition Performed Using Wearable Inertial Sensors Combined With Deep Learning Algorithms

Abstract

This study proposed a wearable device capable of recognizing six human daily activities (walking, walking upstairs, walking downstairs, sitting, standing, and lying) through a deep learning algorithm. Existing wearable devices are mainly watches or wristbands, and almost none are to be worn on the waist. Wearable devices in the forms of watches and wristbands are unfriendly to patients who are critically ill, such as patients undergoing dialysis. Patients undergoing dialysis have artificial blood vessels on their arm, and they cannot perform intense exercise. For this type of users, general hand wearable devices cannot correctly identify wearers' activities. Therefore, we proposed a waist wearable device and these types of daily life activities to assess their exercise. The hardware of the wearable device consisted of an inertial sensor, which included a microcontroller, a three-axis accelerometer, and a three-axis gyroscope. The activity recognition algorithm of the software used motion signals acquisition, signal normalization, and a feature learning method. The feature learning method was based on a 1D convolutional neural network that automatically performed feature extraction and classification from raw data. One part of the experimental data was from the dataset of the University of California (UCI), and the other part was recorded by this study. To capture the data recorded, the wearable inertial sensing device was attached to the waists of 21 experimental participants who performed six common movements in a laboratorial environment, and the subsequent records were collected to verify the validity of the proposed deep learning algorithm in relation to the inertial sensor of the wearable device. For the six common activities in the UCI dataset and the data recorded, the recognition rates in the training sample reached 98.93% and 97.19%, respectively, and the recognition rates in the testing sample were 95.99% and 93.77%, respectively.

ConvAE-LSTM: Convolutional Autoencoder Long Short-Term Memory Network for Smartphone-Based Human Activity Recognition

“…They used the UCI-HAR dataset, a real-world extrasensory dataset [89] and the DCASE 2017 dataset. Yen et al [57] suggested a CNN-based HAR framework using the smartphone-derived UCI-HAR public dataset and selfcollected data from wearable sensors. Wan et al In [59], a sparse AE (SAE) was used to automatically learn features, and based on this concept, a smartphone-based HAR framework was proposed.…”

“…In comparison with a fully connected layer, a convolution layer has much fewer parameters due to sparse connectivity and weight sharing, thereby minimizing the possibility of overfitting. However, due to the tremendous popularity of traditional CNNs in HAR, the recently proposed CNN-based HAR models adopt 1-2 fully connected layers as classifiers [57,87]. Although fully connected layers can adequately perform classification, various parameters lead to the risk of overfitting.…”

Body Temperature Monitoring for Regular COVID-19 Prevention Based on Human Daily Activity Recognition

“…HAR technology using wearable devices and learning algorithms has already achieved several successful applications. Yen et al proposed an automatic CNN feature extraction method combined with a wearable inertial sensor to improve the accuracy of daily activity recognition [4]. Lawal and BanoI developed a method of establishing a frequency image for an original signal to improve the overall recognition performance [10].…”

“…Following the popularization and optimization of intelligent electronic devices and computer systems, human activity recognition (HAR) via wearable devices has been attracting research attention for decades. HAR technology has been applied to various scenarios in the fields of medical care, military security, health service and infotainment, and new application requirements have even emerged in the context of COVID-19 [1][2][3][4][5][6][7][8]. In comparison with vision-based activity recognition, wearable sensors in human activity recognition provide more possibilities for human and mobile devices to interact due to their low cost, small size, strong computing ability and privacy protection, which also serve as the hardware support for the experimental data collection stage.…”

“…Tools such as machine learning (ML) and deep learning (DL) act as core learning algorithms, allowing raw data on human activity recognition to be generalized in various domains after training and testing. Wearable sensing devices typically combine embedded systems with inertial, physiological or environmental sensors to identify ambulation, exercise and daily activities, thus enabling HAR technology not only to perform monitoring and activity prediction but also to provide personalized service and decision support under certain circumstances [4,[9][10][11][12][13][14][15][16][17].…”