Toward practical use of sign language training machine using depth sensor

2017 Fiscal Year Final Research Report

• Project/Area Number: 15K01028
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Educational technology
• Research Institution: Kochi University
• Principal Investigator: Mori Yuichiro 高知大学, 教育研究部自然科学系理学部門, 准教授 (50274361)
• Co-Investigator(Kenkyū-buntansha): 豊永 昌彦 高知大学, 教育研究部自然科学系理学部門, 教授 (40346705)
• Project Period (FY): 2015-04-01 – 2018-03-31
• Keywords: 手話 / トレーニングマシン / 学習支援 / 障害者支援

Outline of Final Research Achievements

This research is the development of Japanese sign language training system that understands, judges and evaluates sign language motion using the original sign language input device in real time.The main results of this research are the following three points.One is an extension of the number of words to be learned. Although it was 231 words in the previous study, we could construct a system that can handle up to the 1000 word level.Furthermore, by improving the judgment section, we improved the influence of individual differences by using fuzzy theory.In addition, the device was changed to Kinect 2, and the accuracy was further improved by the data glove with further improvements, realizing high accuracy and high quality of sign language training.

Free Research Field

ファジィ工学