Human action recognition using wireless wearable in-ear microphone

Jun Nishimura, Tadahiro Kuroda

    Research output: Contribution to journalArticle

    1 Citation (Scopus)

    Abstract

    To realize the ubiquitous eating habits monitoring, we proposed the use of sounds sensed by an in-ear placed wireless wearable microphone. A prototype of wireless wearable in-ear microphone was developed by utilizing a common Bluetooth headset. We proposed a robust chewing action recognition algorithm which consists of two recognition stages: "chew-like" signal detection and chewing sound verification stages. We also provide empirical results on other action recognition using in-ear sound including swallowing, cough, belch, and etc. The average chewing number counting error rate of 1.93% is achieved. Lastly, chewing sound mapping is proposed as a new prototypical approach to provide an additional intuitive feedback on food groups to be able to infer the eating habits in their daily life context.

    Original languageEnglish
    Pages (from-to)1570-1576
    Number of pages7
    JournalIEEJ Transactions on Electronics, Information and Systems
    Volume131
    Issue number9
    DOIs
    Publication statusPublished - 2011

    Keywords

    • Chewing action recognition
    • Eating habits monitoring
    • In-ear microphone
    • In-ear sound recognition

    ASJC Scopus subject areas

    • Electrical and Electronic Engineering

    Fingerprint Dive into the research topics of 'Human action recognition using wireless wearable in-ear microphone'. Together they form a unique fingerprint.

  • Cite this