Novel Design of Interactive Caption and Recognition Postures within the Wearable and Non-Vision System over Wireless Sensor Network

摘要

In this study, we design and present the novel wearable system with the interactive posture caption and recognition functions based on the non-vision over the ZigBee wireless sensor network (ZigBee-WSN). There are two type sensors, 3-axis accelerometer and clip type, to be employed in our wearable system. These sensors are arranged on user's four limbs such that the posture information can be gathered by them. Then, the posture information is transmitted to the data-controlling center over ZigBee-WSN. Finally, this center can analyze and distinguish various postures by our proposed algorithm through the friend user's interface to express. Our presented wearable system can distinguish out 28 kinds of hand postures and 13 kinds of leg postures altogether. Moreover, under our presented ZigBee-WSN system with small size, low-power consumption, and high-reliability characteristics, the data-controlling center can simultaneously tele-monitor the real-time body interactive postures of 8 persons, when the transmitting distance is less than 18 M and the package correct transmitted rate is more than 97.5%.