Classification of Human Motion Activities using Mobile Phone Sensors and Deep Learning Model

Khan, Yusuf Ahmed; Imaduddin, Syed; Prabhat, Ritik; Wajid, Mohd

doi:10.1109/icaccs54159.2022.9785009

Cited by 10 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Mean square errors (MSEs) are often used in regression analysis. But they cannot be used to assess classification problems and can be calculated by squaring the predicted values and the true values [ 19 ].…”

Section: Performance Evaluation and Resultsmentioning

confidence: 99%

“…The Bi-LSTM DNN model uses the data recorded from MEMS sensors either separately or collectively to acquire information like acceleration, magnetic field, orientation, and angular velocity about all three axes (i.e., x, y, and z axes respectively). These MEMS sensors are not only cost-effective but they are also integrated into nearly every smartphone on the market today [ 19 ]. This study covers the following areas and has the following contributions: Rigorous experiments were conducted to prepare an extensive dataset of 9 different human motion activity classes which include ( a ) Laying Down, ( b ) Stationary, ( c ) Walking, ( d ) Brisk Walking, ( e ) Running ( f ) Stairs Up ( g ) Stairs Down ( h ) Squatting and ( i ) Cycling, the prepared dataset was then used for training and testing purposes for the ML and DL model(s).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Khan

Imaduddin

Singh

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

The integration of Micro Electronic Mechanical Systems (MEMS) sensor technology in smartphones has greatly improved the capability for Human Activity Recognition (HAR). By utilizing Machine Learning (ML) techniques and data from these sensors, various human motion activities can be classified. This study performed experiments and compiled a large dataset of nine daily activities, including Laying Down, Stationary, Walking, Brisk Walking, Running, Stairs-Up, Stairs-Down, Squatting, and Cycling. Several ML models, such as Decision Tree Classifier, Random Forest Classifier, K Neighbors Classifier, Multinomial Logistic Regression, Gaussian Naive Bayes, and Support Vector Machine, were trained on sensor data collected from accelerometer, gyroscope, and magnetometer embedded in smartphones and wearable devices. The highest test accuracy of 95% was achieved using the random forest algorithm. Additionally, a custom-built Bidirectional Long-Short-Term Memory (Bi-LSTM) model, a type of Recurrent Neural Network (RNN), was proposed and yielded an improved test accuracy of 98.1%. This approach differs from traditional algorithmic-based human activity detection used in current wearable technologies, resulting in improved accuracy.

show abstract

Section: Performance Evaluation and Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Khan

Imaduddin

Singh

et al. 2023

Sensors

Self Cite

View full text Add to dashboard Cite

show abstract

“…Khan et al [8] have designed a human identification system that can classify human motions based on mobile phone sensors. They used data like accelerometer, Gyroscope, and magnetometer to classify between normal walking and brisk walking.…”

Section: Literature Reviewmentioning

confidence: 99%

Detection of suspicious activity using mobile sensor data and Modified Sub-space K-NN for criminal investigations

Aloni¹,

Shekhawata²

2022

YMER

View full text Add to dashboard Cite

With the bulk availability of mobile sensors, the data collected from them mustn’t be wasted. Nowadays the creation of black-box software that collects this data is not a very difficult task. It is possible to detect suspicious unlawful events using this black-box data. In this paper, we present a novel way of doing forensic investigation using a modified sub-space K-NN (MSK) algorithm. The MSK algorithm is capable of detecting suspicious activities from mobile sensor data. Using this technique, we could detect any normal activity versus suspicious activity with 99.7 % accuracy. We expect the future researcher to develop on this idea and build a solid digital forensic system capable of doing bias-free decisions. Keywords: Forensic, Mobile sensor data, Black box, mobile data collection

show abstract

“…In the paper [ 7 ], the research focused on a hybrid deep learning model that takes heterogeneous sensor data, an acceleration sensor, and an image as inputs. Another work [ 8 ] classified two different physical activities, viz., walking and brisk walking, with deep neural networks from mobile phone sensors such as accelerometers, gyroscopes, magnetometers, etc. Although these approaches can achieve impressive accuracy, deep learning requires a large amount of data from multiple sources to classify activities.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Risk Assessment Detection for Weak People by Parallel Training Logical Execution of a Supervised Learning System Based on an IoT Wearable MEMS Accelerometer

Hoang

Nkembi

Pham

2023

Sensors

View full text Add to dashboard Cite

Activity monitoring has become a necessary demand for weak people to guarantee their safety. The paper proposed a Parallel Training Logical Execution (PTLE) system using machine learning (ML) models on a microelectromechanical system (MEMS) accelerometer to detect coughs, falls, and other normal activities. When there are many categories, the ML prediction can be confused between these activities with each other. The PTLE system trains several models in parallel with more specific activity classes in each dataset. The shared tasks between parallel models relieve the complexity for a single one. There are six additional parameters for accelerometer characteristics, which were calculated from three axes accelerations as input features to improve the ML’s consciousness. Once all models were trained, the system was ready to receive the input accelerations and activated the logical flow to manage link operation between these ML models for output predictions. Random Forest (RF) had the highest potential among the ML classification algorithms after the validation. In the experiment, the comparison between the PTLE model and the regular ML model were carried out with real-time data from an M5stickC wearable device on the user’s chest to the trained models on PC. The result showed the advancement of the proposed method in term of precision, recall, F1-score with an overall accuracy of 98% in the real-time test. The accelerations from the wearable device were sent to ML models via Wi-Fi with Message Queue Telemetry Transport (MQTT) broker, and the activity predictions were transferred to the cloud for the family members or doctor care based on Internet of Things (IoT) communication.

show abstract

Classification of Human Motion Activities using Mobile Phone Sensors and Deep Learning Model

Cited by 10 publications

References 12 publications

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Artificial Intelligence Based Approach for Classification of Human Activities Using MEMS Sensors Data

Detection of suspicious activity using mobile sensor data and Modified Sub-space K-NN for criminal investigations

Real-Time Risk Assessment Detection for Weak People by Parallel Training Logical Execution of a Supervised Learning System Based on an IoT Wearable MEMS Accelerometer

Contact Info

Product

Resources

About