ロボットハンド制御のための手指形状推定(「手」及びヒューマン情報処理一般)

概要

論文の詳細を見る
The processing time for human hand posture estimation should be equal for the purpose of stable control of robots. The hand posture estimation by means of searching a similar image from a vast database, moreover, may cause the increase of processing time, and prevent realtime controlling of a robot. In this study, therefore, the author proposed a new estimation method of human hand posture by rearranging the vast database with the Self-Organizing Map (SOM) including self-reproduction and self-annihilation, which enables two-step searches of similar image with short and equal period of processing time within small errors. The experimental results showed that our the system exhibited good performance with high accuracy within processing time of 20ms or less for each image input with a 2.8GHz CPU PC.
社団法人電子情報通信学会の論文
2005-10-13