Motion Sensitive Glove-Based Korean Fingerspelling Tutor
Yunli Lee, Seungki Min, Hwang-Kyu Yang, Keechul Jung · 2007 International Conference on Convergence Information Technology (ICCIT 2007) · 2007
Korean fingerspelling serves as a more cost effective approach in designing data glove for fingerspelling recognition purpose. This motivates us to develop a system that provides learners with a self learning module and has their own personal tutor - data glove. Therefore, in this paper, we introduce a motion sensitive glove-based Korean Fingerspelling Tutor (KFT) using our invented data glove. The data glove only needs 2 tilt sensors, 5 flex sensors and 3 pressure sensors in order to support motion sensitive glove-based KFT at a reasonable cost. This system achieves an average of accuracy above 80% for the 24 letters of Korean fingerspelling. The KFT contains three learning modules: letters, words, and short sentences, and a fingerspelling game, that both offers effective learning environment and hands-on practice experiences to the learner. This system provides a tremendously fast and effective learning process of Korean fingerspelling throughout the personal tutor - data glove.