scholarly journals Real-Time Physical Activity Recognition on Smart Mobile Devices Using Convolutional Neural Networks

2020 ◽  
Vol 10 (23) ◽  
pp. 8482
Author(s):  
Konstantinos Peppas ◽  
Apostolos C. Tsolakis ◽  
Stelios Krinidis ◽  
Dimitrios Tzovaras
Keyword(s):  
Neural Network ◽  
Physical Activity ◽  
Real Time ◽  
Mobile Devices ◽  
Mobile Device ◽  
Classification Performance ◽  
Sensor Data ◽  
Accelerometer Data ◽  
Smart Mobile ◽  
Fully Connected

Given the ubiquity of mobile devices, understanding the context of human activity with non-intrusive solutions is of great value. A novel deep neural network model is proposed, which combines feature extraction and convolutional layers, able to recognize human physical activity in real-time from tri-axial accelerometer data when run on a mobile device. It uses a two-layer convolutional neural network to extract local features, which are combined with 40 statistical features and are fed to a fully-connected layer. It improves the classification performance, while it takes up 5–8 times less storage space and outputs more than double the throughput of the current state-of-the-art user-independent implementation on the Wireless Sensor Data Mining (WISDM) dataset. It achieves 94.18% classification accuracy on a 10-fold user-independent cross-validation of the WISDM dataset. The model is further tested on the Actitracker dataset, achieving 79.12% accuracy, while the size and throughput of the model are evaluated on a mobile device.

scholarly journals Using Accelerometer and GPS Data for Real-Life Physical Activity Type Detection

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 588 ◽  
Author(s):  
Hoda Allahbakhshi ◽  
Lindsey Conrow ◽  
Babak Naimi ◽  
Robert Weibel
Keyword(s):  
Physical Activity ◽  
Real Life ◽  
Classification Performance ◽  
Training Data ◽  
Sensor Data ◽  
Gps Data ◽  
Accelerometer Data ◽  
Life Data ◽  
Sensor Position ◽  
Real Life Data

This paper aims to examine the role of global positioning system (GPS) sensor data in real-life physical activity (PA) type detection. Thirty-three young participants wore devices including GPS and accelerometer sensors on five body positions and performed daily PAs in two protocols, namely semi-structured and real-life. One general random forest (RF) model integrating data from all sensors and five individual RF models using data from each sensor position were trained using semi-structured (Scenario 1) and combined (semi-structured + real-life) data (Scenario 2). The results showed that in general, adding GPS features (speed and elevation difference) to accelerometer data improves classification performance particularly for detecting non-level and level walking. Assessing the transferability of the models on real-life data showed that models from Scenario 2 are strongly transferable, particularly when adding GPS data to the training data. Comparing individual models indicated that knee-models provide comparable classification performance (above 80%) to general models in both scenarios. In conclusion, adding GPS data improves real-life PA type classification performance if combined data are used for training the model. Moreover, the knee-model provides the minimal device configuration with reliable accuracy for detecting real-life PA types.

scholarly journals Optimization of Physical Activity Recognition for Real-Time Wearable Systems: Effect of Window Length, Sampling Frequency and Number of Features

2019 ◽  
Vol 9 (22) ◽  
pp. 4833 ◽  
Author(s):  
Ardo Allik ◽  
Kristjan Pilt ◽  
Deniss Karai ◽  
Ivo Fridolin ◽  
Mairo Leier ◽  
...  
Keyword(s):  
Physical Activity ◽  
Feature Selection ◽  
Real Time ◽  
Sampling Frequency ◽  
Classification Performance ◽  
Window Length ◽  
Decision Tree Classifier ◽  
Acceleration Signal ◽  
Wearable Systems ◽  
Tree Classifier

The aim of this study was to develop an optimized physical activity classifier for real-time wearable systems with the focus on reducing the requirements on device power consumption and memory buffer. Classification parameters evaluated in this study were the sampling frequency of the acceleration signal, window length of the classification fragment, and the number of classification features, found with different feature selection methods. For parameter evaluation, a decision tree classifier was created based on the acceleration signals recorded during tests, where 25 healthy test subjects performed various physical activities. Overall average F1-score achieved in this study was about 0.90. Similar F1-scores were achieved with the evaluated window lengths of 5 s (0.92 ± 0.02) and 3 s (0.91 ± 0.02), while classification performance with 1 s were lower (0.87 ± 0.02). Tested sampling frequencies of 50 Hz, 25 Hz, and 13 Hz had similar results with most classified activity types, with an exception of outdoor cycling, where differences were significant. Using forward sequential feature selection enabled the decreasing of the number of features from initial 110 features to about 12 features without lowering the classification performance. The results of this study have been used for developing more efficient real-time physical activity classifiers.

scholarly journals A Smart Collaborative Policy for Mobile Fog Computing in Rural Vitalization

2018 ◽  
Vol 2018 ◽  
pp. 1-10 ◽  
Author(s):  
Yutong Zhou ◽  
Wei Shi ◽  
Fei Song
Keyword(s):  
Cloud Computing ◽  
Real Time ◽  
Mobile Devices ◽  
Policy Design ◽  
Fog Computing ◽  
Performance Evaluations ◽  
Round Trip Time ◽  
Round Trip ◽  
Trip Time ◽  
Smart Mobile

Mobile Fog Computing (MFC), as a crucial supplement to cloud computing, has its own special traits in many aspects. As smart mobile devices grow and vary in shapes and formats over the years, the need for real-time interactions and an easy-to-use network is imminent. In this paper, we propose a smart collaborative policy for MFC scenarios by considering the target of rural vitalization. The challenges and drawbacks of extending cloud to fog are reviewed at the beginning. Then, the analysis of policy design is presented from the perspectives of feature comparisons, urgent requirements, and possible solutions. The details of policy establishment are introduced with necessary examples. Finally, performance evaluations are provided based on simulation platforms. Validation results related to round trip time and transmission time illustrate the significant improvements of our proposal in certain ways compared to the original candidate, which enables larger deployment in impoverished areas.

Know Your World Better

2018 ◽  
pp. 777-793
Author(s):  
Srinivasa K. G. ◽  
Satvik Jagannath ◽  
Aakash Nidhi
Keyword(s):  
Information Retrieval ◽  
Augmented Reality ◽  
Real Time ◽  
Mobile Devices ◽  
Mobile Device ◽  
Sensitive Information ◽  
Context Sensitive ◽  
Close Proximity ◽  
Augmented Reality Application ◽  
Device User

Mobile devices are changing the way people live. Users have everything on their fingertips and to support them, there are scores of application which add to the usability and comfort. “Know your world better” is an Augmented Reality application developed for Android. This application helps the user to find friends and locate places in close proximity. In this paper we talk about an application that describes a method of augmenting Point of Interests (POI's) on a mobile device. User has to move his phone pointing in a direction of his choice and POI's if any are shown in real time. The user's interest with respect to the environment is inferred from speech or by selecting from the choices; this data is used for information retrieval from the cloud. The result of context-sensitive information retrieval is augmented onto the view of the mobile and provides speech output.

scholarly journals A Karaoke System with Real-Time Media Merging and Sharing Functions for a Cloud-Computing-Integrated Mobile Device

2013 ◽  
Vol 2013 ◽  
pp. 1-8
Author(s):  
Her-Tyan Yeh ◽  
Juing-Shian Chiou ◽  
Ting-Jun Zhou
Keyword(s):  
Cloud Computing ◽  
Real Time ◽  
Mobile Devices ◽  
Mobile Device ◽  
Personal Digital Assistants ◽  
Popular Entertainment ◽  
Work Related ◽  
Cloud Server ◽  
Streaming Technology ◽  
Technology Users

Mobile devices such as personal digital assistants (PDAs), smartphones, and tablets have increased in popularity and are extremely efficient for work-related, social, and entertainment uses. Popular entertainment services have also attracted substantial attention. Thus, relevant industries have exerted considerable efforts in establishing a method by which mobile devices can be used to develop excellent and convenient entertainment services. Because cloud-computing technology is mature and possesses a strong computing processing capacity, integrating this technology into the entertainment service function in mobile devices can reduce the data load on a system and maintain mobile device performances. This study combines cloud computing with a mobile device to design a karaoke system that contains real-time media merging and sharing functions. This system enables users to download music videos (MVs) from their mobile device and sing and record their singing by using the device. They can upload the recorded song to the cloud server where it is merged with real-time media. Subsequently, by employing a media streaming technology, users can store their personal MVs in their mobile device or computer and instantaneously share these videos with others on the Internet. Through this process, people can instantly watch shared videos, enjoy the leisure and entertainment effects of mobile devices, and satisfy their desire for singing.

scholarly journals The Effectiveness of Mobile Device-Based Digital Interventions on the Risk Factors of Diabetes Mellitus Control in the Industrial Revolution 4.0

2021 ◽  
Vol 5 (4) ◽  
pp. 320
Author(s):  
Ratna Kuatiningsari ◽  
Fatqiatul Wulandari ◽  
Ade Lia Ramadani ◽  
Qonita Rachmah
Keyword(s):  
Physical Activity ◽  
Diabetes Mellitus ◽  
Risk Factors ◽  
Mobile Devices ◽  
Mobile Device ◽  
Self Efficacy ◽  
Industrial Revolution ◽  
Self Management ◽  
Original Research ◽  
Hba1c Levels

ABSTRACTBackground: Diabetes mellitus is a chronic disease which if not done properly, can cause microvascular and macrovascular disorders. Indicators of the accuracy of diabetes management in this scientific article include education, self-management (improving diet, increasing physical activity, and self-efficacy), and monitoring of HbA1c levels. Mobile devices have the potential as a tool for diabetes mellitus management in the era of the industrial revolution 4.0.Purpose: to provide the latest information regarding the effectiveness of using mobile devices in controlling risk factors for diabetes mellitus.Method: This study is a literature review study. The electronic databases used are Google Scholar, Science Direct, and Directory of Access Journals (DOAJ). Inclusion criteria: original research, a journal of at least 80% indexed by Sinta (Indonesian journal) and indexed by Scopus (international journal), published year 2010-2020, intervention using a mobile device, has an output of HbA1c levels, self management (diet, physical activity, and self efficacy), and the level of knowledge. Exclusion criteria: reference with secondary data.Result: This study used 16 scientific articles. A number of 12 studies (75%) reported the use of mobile device applications in controlling risk factors for diabetes mellitus had significant measurement results in controlling HbA1c levels in 10 studies (83%) and 2 studies were not significant (17%). Outcomes in the form of self-management were reported by 9 studies with details of the significant results of dietary improvement in 5 studies (83%), increased physical activity in 5 studies (63%), and self-efficacy in 4 studies (67%). The increase in knowledge was reported by 4 studies with significant results (100%).Conclusion: Mobile device-based digital intervention is quite effective in controlling diabetes mellitus risk factors to control HbA1c levels, increasing self-management (improving diet, increasing physical activity, and self-efficacy) and knowledge. 

scholarly journals Can Pre-Trained Convolutional Neural Networks be directly used as a Feature Extractor for Video-based Neonatal Sleep and Wake Classification?

2020 ◽  
Author(s):  
Muhammad Awais ◽  
Xi Long ◽  
Bin Yin ◽  
Chen chen ◽  
Saeed Akbarzadeh ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Convolutional Neural Networks ◽  
Video Recording ◽  
Principal Component ◽  
Classification Performance ◽  
Support Vector ◽  
Video Frames ◽  
Feature Extractor ◽  
Fully Connected

Abstract Objective: In this paper, we propose to evaluate the use of a pre-trained convolutional neural networks (CNNs) as a features extractor followed by the Principal Component Analysis (PCA) to find the best discriminant features to perform classification using support vector machine (SVM) algorithm for neonatal sleep and wake states using Fluke® facial video frames. Using pre-trained CNNs as feature extractor would hugely reduce the effort of collecting new neonatal data for training a neural network which could be computationally very expensive. The features are extracted after fully connected layers (FCL’s), where we compare several pre-trained CNNs, e.g., VGG16, VGG19, InceptionV3, GoogLeNet, ResNet, and AlexNet. Results: From around 2-h Fluke® video recording of seven neonate, we achieved a modest classification performance with an accuracy, sensitivity, and specificity of 65.3%, 69.8%, 61.0%, respectively with AlexNet using Fluke® (RGB) video frames. This indicates that using a pre-trained model as a feature extractor could not fully suffice for highly reliable sleep and wake classification in neonates. Therefore, in future a dedicated neural network trained on neonatal data or a transfer learning approach is required.

Sign in / Sign up

Export Citation Format

Share Document