3703 三軸加速度情報を用いた人間の行動判別に基づく文脈推定手法の研究(S55-1 生物医学工学における計測と制御(1),21世紀地球環境革命の機械工学:人・マイクロナノ・エネルギー・環境)

概要

論文の詳細を見る
Recently developed classification schemes of human actions consider a limited number of actions and are therefore not adequate for modeling an entire daily life pattern. This paper presents a context-aware classification method of human actions, which allows for a large number of actions to be classified. The proposed method makes it possible to divide human actions into an adequate number of activities without .a priori knowledge of this number by using hierarchical clustering of 3-dimensional acceleration sensor and electrical potential sensor data. We defined a new index, the predicted information amount, that described the context. It was useful for determination of the optimum number of clusters used in the classification method. The proposed index was successfully applied in the discrimination of the actual context of situations encountered in daily life.
社団法人日本機械学会の論文
2008-08-02