We explored the effects of mandala art therapy game for attention deficit children who were enrolled in day care center for children. We found 17 ADD children through questionnaire DSM-Ⅳ. For 2 Months, from 1st of April to 20th of May, experiments of art therapy program and Mandala art therapy games were carried out twice a week. After analysis of experimental results, Mandala art therapy game is effective for ADD children.

In this paper we offered a new UI design for Chinese input method on smart devices during Gameplay. Currently the well-known conventional stroke-based Chinese input method using only five basic stroke types could achieve low leaning curve and small keypad implementation, we made the UI design base on this stroke-based input method compensating the problem that input speed is limited and finally relatively higher the input speed. For experiment we designed and implemented the new UI for input method by Unity3D. Input by pressing or typing keys is replaced by more using sliding. To improve the input speed of the stroke-based method slide direction arrows are used and selected character will be intuitively shown to user in real time as designed. And the Four Tones(四声) in Chinese pronunciation is also been used to improve the input speed for this input method. In addition, to evaluation the new input method UI, we have conducted experiments on smart device comparing with four basic Chinese input methods. Through the numeric values recorded in the evaluation by inputting random 20 Chinese characters and two-letter Chinese words, comparing the input speed, click times, the typo rate shows the advantages of the new UI and proved that the new UI is better than others for using in smart device game.

최근 키보드, 마우스, 조이스틱 등과 같은 게임 입력장치의 과도한 사용에 따른 부작용으로 손목 터널증후군을 호소하는 인구가 늘어나고 있다. 그러므로 본 논문에서는 생체신호를 게임의 입력장치로 사용하는 방법을 제안한다. 손목의 움직임시 발생되는 근전도 신호를 입력받아 효과적으로 잡음을 제거하고 신호를 증폭하는 하드웨어 모듈을 설계하였다. 또한, 웨이브릿 패킷 변환을 통하여 동작의 특징벡터를 추출하고, 주성분 분석을 통하여 특징벡터를 재배치함으로서 특징벡터의 분산을 감소시켰다. 데이터 분류 알고리즘인 SVM을 적용하여 학습된 데이터를 구축하였으며, 4가지 동작에 대해 90% 이상의 인식률을 나타내었다.