Publication: An integration of Myo armbands and an android-based mobile application for communication with hearing-impaired persons
Issued Date
2018-04-09
Resource Type
Other identifier(s)
2-s2.0-85048864804
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
Proceedings - 13th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2017. Vol.2018-January, (2018), 413-417
Suggested Citation
Malika Vachirapipop, Safra Soymat, Wasurat Tiraronnakul, Narit Hnoohom An integration of Myo armbands and an android-based mobile application for communication with hearing-impaired persons. Proceedings - 13th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2017. Vol.2018-January, (2018), 413-417. doi:10.1109/SITIS.2017.74 Retrieved from: https://repository.li.mahidol.ac.th/handle/123456789/45640
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
An integration of Myo armbands and an android-based mobile application for communication with hearing-impaired persons
Other Contributor(s)
Abstract
© 2017 IEEE. Hearing-impaired people or those with any other disabilities lack support and help. For the hearing-impaired, communication to the rest of the world is limited due to the limited number of interpreters. With our knowledge, this problem has motivated the researchers to create a medium to support these people and provide them with the ability to communicate freely in different situations. This paper includes two main actors; namely, Myo armbands and a mobile application. The Myo armbands capture muscular movements and send the captured frequency as an input to the Android-based mobile application in which then, through the embedded prediction model, the mapping of the input data occurs. Once the translation was completed, it was sent back to the mobile screen where the translation was displayed. The application was built to translate a total of six gestures, where they were then classified into three categories; namely, daily communication, illness and emergency situations. The accuracy of the application for being able to translate the gestures into the correct meaning was tested by 12 users and resulted in the accuracy of five out of six signs. The result has, however, shown some gestures to be confused with others. The gestures that were seen confused with each other was sorry and help. The reason for this confusion can be concluded in two main points. The first being too few data sets that were used for training, and second being the gestures had a close posture, i.e. the position, height and orientation of the hands. This problem was, however, able to be solved by gesture performance guidance.