Accelerometer based human activities and posture recognition

Arun Babu*, Kudakwashe Dube, Subhas Mukhopadhyay, Hemant Ghayvat, M. V. Jithin Kumar

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contributionpeer-review

6 Citations (Scopus)


The quantity of elderly people like to live in their homes, secluded, in their brilliant age is expanding exponentially. This is not a perfect path for an elderly individual to live. However, the urbanization and resultant change of the social and social conduct makes it a more regular event. Falls are a noteworthy reason for death and horribleness in more established grown-ups. In this way, it has turn into an opportune need to create mechanized look after the elderly. The first end, purpose of such fall computer looking-glass is to ready caregivers of the fall event, which can then start an earlier process. In the present study, we amplify the use of wearable inertial sensors for fall identification and information of human posture and activities, by creating and assessing the precision of a sensor framework for identifying the same. We found that our system could discover fall events and monitor human activities with at least 95% accuracy.

Original languageEnglish
Title of host publicationSAPIENCE 2016
Subtitle of host publicationProceedings of 2016 International Conference on Data Mining and Advanced Computing
Place of PublicationPiscataway, NJ
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Number of pages7
ISBN (Electronic)9781467385947
Publication statusPublished - 25 Oct 2016
Externally publishedYes
Event2016 International Conference on Data Mining and Advanced Computing, SAPIENCE 2016 - Ernakulam, India
Duration: 16 Mar 201618 Mar 2016


Other2016 International Conference on Data Mining and Advanced Computing, SAPIENCE 2016


  • fall detection
  • wearable sensors
  • accelerometer
  • posture recognition
  • human activities


Dive into the research topics of 'Accelerometer based human activities and posture recognition'. Together they form a unique fingerprint.

Cite this