Design of Optimal Deep Learning Based Human Activity Recognition on Sensor Enabled Internet of Things Environment

Fahd N Al-Wesabi,Saleh Alhazbi,Manar Ahmed Hamza,Asma Abdulghani Al-Shargabi,Amani Abdulrahman Albraikan,Mesfer Al Duhayyim,Anwer Mustafa Hilal,Mohammed Rizwanullah

doi:10.1109/access.2021.3112973

Abstract

In recent times, Human activity recognition (HAR) becomes a major challenging issue among computer vision applications in day to day lives. The HAR is mainly envisioned to be utilized with other technologies, namely Internet of Things (IoT) and sensor technologies. Due to the advancements of deep learning (DL) approaches, automated high level feature extraction process can be utilized to improve the HAR results. In addition, DL techniques can be employed in different domains of sensor enabled HAR. In this aspect, this study designs an optimal DL based HAR (ODL-HAR) model on sensor enabled IoT environments. The ODL-HAR technique aims to determine human activities in day to day lives using wearables and IoT devices. The proposed ODL-HAR technique involves different stages of operations namely data acquisition, data preprocessing, feature extraction, classification, and parameter optimization. The ODL-HAR technique uses a MobileNet-v2 model as a feature extractor and bidirectional long short-term memory (BiLSTM) model as a classifier. In order to optimally tune the hyperparameter involved in the BiLSTM model, chaos game optimization (CGO) algorithm is employed and thereby raises the recognition performance. The design of CGO algorithm for hyperparameter optimization of HAR shows the novelty of the work. A wide range of simulations takes place to point out the betterment of the ODL-HAR technique on two benchmark datasets. The experimental results portrayed the enhanced performance of the ODL-HAR technique over the other recent HAR approaches interms of different evaluation parameters.

Highlights

Nowadays, human activity recognition (HAR) has become a significant part of human lives
The results demonstrated that the optimal DL based HAR (ODL-HAR) technique has proficiently identified the class labels under all sizes of training/testing data
The ROC analysis of the ODL-HAR techniques take place on the UCI HAR dataset is depicted in Fig. 5 under different sizes of training and testing data

Summary

Introduction

Human activity recognition (HAR) has become a significant part of human lives It identifies the humans’ current/presence activity or action or state which is based on the data i.e., being obtained from the different sensor nodes. These systems are employed for detecting the humans’ movement day-to-day actions. A HAR method has wide range of applications, like remote monitoring, healthcare, military, sports, gaming, surveillance, security, interactions between computers and humans. This support scheme has another wide range of applications, like monitoring sleep disorders, humans’ special activity, fall detection, rehabilitation activity recognition, and so on. For mental and physical development, physical activity plays an important part that

Methods

Results

Conclusion