IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
Hi-index | 0.00 |
An extension limb action recognition method based on acceleration median (EULAR-AM) is proposed in this paper. Stretch arm has the feature that the arm's acceleration increases firstly, and then decreases. So the EULAR-AM chooses the acceleration median of the arm outstretching process and the direction of acceleration at the initial moment of arm outstretching as its recognition characteristic values. It can reduce the affection of outstretching speed to the characteristic value of limb action, and can achieve the goal that the different outstretching speed actions having same direction could be described by the same characters. Then combining the extension recognition method, the EULAR-AM recognized the limb action. The experiment results show that the recognition accuracy rate of the EULAR-AM is 93.2 %.