Show simple item record

dc.contributor.authorVafeiadis, Anastasiosen
dc.contributor.authorVotis, Konstantinosen
dc.contributor.authorGiakoumis, Dimitriosen
dc.contributor.authorTzovaras, Dimitriosen
dc.contributor.authorChen, Limingen
dc.contributor.authorHamzaoui, Raoufen
dc.date.accessioned2017-06-23T12:22:57Z
dc.date.available2017-06-23T12:22:57Z
dc.date.issued2017
dc.identifier.citationVafeiadis, A., Votis, K., Giakoumis, D., Tzovaras, D., Chen, L., Hamzaoui, R. (2017) Audio-based event recognition system for smart homes. In: Proc. 14th IEEE International Conference on Ubiquitous Intelligence and Computing, San Francisco, CA, Aug. 2017.en
dc.identifier.urihttp://hdl.handle.net/2086/14256
dc.description.abstractBuilding an acoustic-based event recognition system for smart homes is a challenging task due to the lack of high-level structures in environmental sounds. In particular, the selection of effective features is still an open problem. We make an important step toward this goal by showing that the combination of Mel-Frequency Cepstral Coefficients, Zero- Crossing Rate, and Discrete Wavelet Transform features can achieve an F1 score of 96.5% and a recognition accuracy of 97.8% with a gradient boosting classifier for ambient sounds recorded in a kitchen environment.en
dc.language.isoen_USen
dc.publisherIEEEen
dc.subjectSmart homesen
dc.subjectAssisted livingen
dc.subjectActivity recognitionen
dc.subjectAudio feature extractionen
dc.subjectClassificationen
dc.titleAudio-based Event Recognition System for Smart Homesen
dc.typeConferenceen
dc.researchgroupCIIRGen
dc.peerreviewedYesen
dc.explorer.multimediaNoen
dc.funderEUen
dc.projectidACROSSINGen
dc.projectid676157en
dc.cclicenceCC-BY-NCen
dc.date.acceptance2017-05-10en
dc.researchinstituteCyber Technology Institute (CTI)en
dc.researchinstituteInstitute of Artificial Intelligence (IAI)en
dc.researchinstituteInstitute of Engineering Sciences (IES)en


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record