Development of Finger Braille Recognition System

Finger Braille is one of the communication media of deafblind people. In one-handed Finger Braille, a sender dots the left part of the Braille code on the distal interphalangeal (DIP) joints of the index, middle and ring fingers of a receiver, and subsequently dots the right part of the Braille code...

Full description

Bibliographic Details
Main Authors: Yasuhiro MATSUDA, Ichiro SAKUMA, Yasuhiko JIMBO, Etsuko KOBAYASHI, Tatsuhiko ARAFUNE, Tsuneshi ISOMURA
Format: Article
Language:English
Published: The Japan Society of Mechanical Engineers 2010-03-01
Series:Journal of Biomechanical Science and Engineering
Subjects:
Online Access:https://www.jstage.jst.go.jp/article/jbse/5/1/5_1_54/_pdf/-char/en
Description
Summary:Finger Braille is one of the communication media of deafblind people. In one-handed Finger Braille, a sender dots the left part of the Braille code on the distal interphalangeal (DIP) joints of the index, middle and ring fingers of a receiver, and subsequently dots the right part of the Braille code on the proximal interphalangeal (PIP) joints of the same fingers. Because there is a small number of non-disabled people who are skilled in Finger Braille, deafblind people communicate in this medium only through an interpreter. In this study, we developed a Finger Braille recognition system using small piezoelectric accelerometers worn by the receiver. We first conducted a measurement experiment to derive algorithms for the recognition of the dotted fingers and positions. The results showed that the accuracy of the recognition of dotted fingers was 92.9%, and the recognition system was independent of the dotted strength, dotted position and receiver. The accuracy of the recognition of dotted positions was 81.9%, and the recognition system could recognize dotted positions if the receiver's hand formed a natural longitudinal arch on the desk. Next, an evaluation experiment was carried out. The results showed that the accuracy of the recognition of dotted fingers by dotting was 89.7%, and the accuracy of recognition of dotted positions by dotting was 92.3%. Therefore, the recognition system could recognize sentences accurately when the interpreter dotted clearly.
ISSN:1880-9863