A novel feature extraction technique for human activity recognition

Víctor Elvira, Alfredo Nazábal-Rentería, Antonio Artés-Rodríguez

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract / Description of output

This work presents a novel feature extraction technique for human activity recognition using inertial and magnetic sensors. The proposed method estimates the orientation of the person with respect to the earth frame by using quaternion representation. This estimation is performed automatically without any extra information about where the sensor is placed on the body of the person. Furthermore, the method is also robust to displacements of the sensor with respect to the body. This novel feature extraction technique is used to feed a classification algorithm showing excellent results that outperform those obtained by an existing state-of-the-art feature extraction technique.

Original languageEnglish
Title of host publication2014 IEEE Workshop on Statistical Signal Processing, SSP 2014
PublisherIEEE Computer Society
Pages177-180
Number of pages4
ISBN (Print)9781479949755
DOIs
Publication statusPublished - 1 Jan 2014
Event2014 IEEE Workshop on Statistical Signal Processing, SSP 2014 - Gold Coast, QLD, Australia
Duration: 29 Jun 20142 Jul 2014

Conference

Conference2014 IEEE Workshop on Statistical Signal Processing, SSP 2014
Country/TerritoryAustralia
CityGold Coast, QLD
Period29/06/142/07/14

Keywords / Materials (for Non-textual outputs)

  • Activity Classification
  • Ambulatory Monitoring
  • Features Extraction
  • Inertial Sensors
  • Magnetic Sensors
  • Orientation Estimation
  • Quaternions

Fingerprint

Dive into the research topics of 'A novel feature extraction technique for human activity recognition'. Together they form a unique fingerprint.

Cite this