A smartphone-based real-time simple activity recognition

Jongprasithporn, Manutchanok; Yodpijit, Nantakrit; Srivilai, Rawiphorn; Pongsophane, Paweena

doi:10.1109/iccar.2017.7942755

Cited by 7 publications

(20 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The comparison in Table 6 allows us to locate the present work with an overall accuracy that is only surpassed by a system that only considers three activities and by our previous work in [26]. However, the only work from Table 6 that displays results of real-time tests is [30]. There, tests are 10 s long, compared to the 30 s tests of the present work.…”

Section: Discussionmentioning

confidence: 91%

“…There, tests are 10 s long, compared to the 30 s tests of the present work. Reference [30] considers six test subjects, while our case study considered twenty. Hence, such a length and subject quantity difference can lead to errors in movement data, which makes the results shown in Table 4 generalizable to expected performance during real use.…”

Section: Discussionmentioning

confidence: 99%

“…The integration device, typically a smartphone, can take the place of a movement sensor [29]. Accordingly, [30][31][32] describe smartphone-based HAR systems along with their corresponding challenges regarding feature extraction and selection. However, these systems are highly dependent on the on-body location of the smartphone.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Physical Workload Tracking Using Human Activity Recognition with Wearable Devices

Manjarrés

Narváez

Gasser

et al. 2019

Sensors

View full text Add to dashboard Cite

In this work, authors address workload computation combining human activity recognition and heart rate measurements to establish a scalable framework for health at work and fitness-related applications. The proposed architecture consists of two wearable sensors: one for motion, and another for heart rate. The system employs machine learning algorithms to determine the activity performed by a user, and takes a concept from ergonomics, the Frimat's score, to compute the corresponding physical workload from measured heart rate values providing in addition a qualitative description of the workload. A random forest activity classifier is trained and validated with data from nine subjects, achieving an accuracy of 97.5%. Then, tests with 20 subjects show the reliability of the activity classifier, which keeps an accuracy up to 92% during real-time testing. Additionally, a single-subject twenty-day physical workload tracking case study evinces the system capabilities to detect body adaptation to a custom exercise routine. The proposed system enables remote and multi-user workload monitoring, which facilitates the job for experts in ergonomics and workplace health.

show abstract

Section: Discussionmentioning

confidence: 91%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Physical Workload Tracking Using Human Activity Recognition with Wearable Devices

Manjarrés

Narváez

Gasser

et al. 2019

Sensors

View full text Add to dashboard Cite

show abstract

“…In summary, articles were excluded for not examining the methodological effectiveness of the measurement approach, not reporting on at least one of the 'big six' [15] or on feasibility, and not using a smartphone (Figure 1). When applying the inclusion criteria for participants to be healthy young people, only 5 studies [23][24][25][26][27] were eligible. The key characteristics of the included study populations are presented in Table 2.…”

Section: Summary Of Included Studiesmentioning

confidence: 99%

Measurement Properties of Smartphone Approaches to Assess Physical Activity in Healthy Young People: Systematic Review

Parmenter¹,

Burley²,

Stewart³

et al. 2022

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

Background Physical inactivity is a preventable risk factor for several chronic diseases and one of the driving forces behind the growing global burden of disease. Recent evidence has shown that interventions using mobile smartphone apps can promote a significant increase in physical activity (PA) levels. However, the accuracy and reliability of using apps is unknown. Objective The aim of our review was to determine the accuracy and reliability of using mobile apps to measure PA levels in young people. We conducted a systematic review guided by PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). Methods Studies published from 2007 to 2020 were sourced from 8 databases—Ovid MEDLINE, Embase (Elsevier), Cochrane Library (Wiley), PsychINFO (EBSCOhost), CINAHL (EBSCOhost), Web of Science (Clarivate), SPORTDiscus (EBSCOhost), and IEEE Xplore Digital Library database. Studies were conducted in young people aged 10-24 years and without chronic illnesses, who evaluated a mobile app’s ability to measure PA. Primary outcomes included validity, reliability, and responsiveness of the measurement approach. Duplicate screening was conducted for eligibility, data extraction, and assessing the risk of bias. Results were reported as a systematic review. The main physical activity measures evaluated for each study were the following: total PA time (min/day or min/week), total moderate to vigorous PA per week, daily step count, intensity measure (heart rate), and frequency measure (days per week). Results Of the 149 identified studies, 5 met the inclusion criteria (322 participants, 176 female; mean age 14, SD 3 years). A total of 3 studies measured criterion validity and compared PA measured via apps against PA measured via an Actigraph accelerometer. The 2 studies that reported on construct validity identified a significant difference between self-reported PA and the objective measure. Only 1 of the 5 apps examined was available to the public, and although this app was highly accepted by young people, the app recorded PA to be significantly different to participants’ self-reported PA. Conclusions Overall, few studies assess the reliability, validity, and responsiveness of mobile apps to measure PA in healthy young people, with studies typically only reporting on one measurement property. Of the 3 studies that measured validity, all concluded that mobile phones were acceptable and valid tools. More research is needed into the validity and reliability of smartphone apps to measure PA levels in this population as well as in populations with other characteristics, including other age groups and those with chronic diseases. Trial Registration PROSPERO CRD42019122242; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=122242

show abstract

“…Literature [21] used wrist motion sensor and built-in sensors in mobile phones to collect the experimenters' movement data, identified two sets of a total of 13 kinds of complex movements, but as the system added additional sensors, the user costs have been largely increased. Literature [13] used the standard deviation of the linear acceleration sensor as feature vector to identify standing, walking and running. Although the accuracy is more than 98%, the recognized motion states are very few.…”

Section: Related Studiesmentioning

confidence: 99%

Real-Time Recognition of Human Daily Motion with Smartphone Sensor

Qi-shou¹

2018

IJPE

View full text Add to dashboard Cite

Aiming at problems regarding the recognition of motion states by existing smartphones, such as poor real-time performance, less movement category and complex algorithm, this paper proposes a method of using smartphone sensors to recognize six kinds of real time human movement states. Firstly, daily human movement data is acquired through smartphone acceleration sensors and gravitational acceleration sensors, and original data is handled with correction, smoothing, segmentation and direction-independent processing. Secondly, the footsteps identification algorithm is used to calculate peaks and troughs of footsteps from which the time-domain feature vectors are extracted. Finally, the movement states are classified according to feature vectors, and the Hierarchical Support Machines (H-SVMs) is used to recognize daily movement states. Experimental results show this method can effectively reduce the computational load of smartphones and improve real-time performance and accuracy of movement states recognition. This method is suitable for other similar behavior recognitions.

show abstract

A smartphone-based real-time simple activity recognition

Cited by 7 publications

References 12 publications

Physical Workload Tracking Using Human Activity Recognition with Wearable Devices

Physical Workload Tracking Using Human Activity Recognition with Wearable Devices

Measurement Properties of Smartphone Approaches to Assess Physical Activity in Healthy Young People: Systematic Review

Real-Time Recognition of Human Daily Motion with Smartphone Sensor

Contact Info

Product

Resources

About