基于Kinect的动态手势识别研究

针对人机交互领域中基于视觉的传统动态手势识别方法准确率不高、易受不同强度光照影响等问题,对动态手势识别方法进行了研究;首先利用Kinect传感器采集的深度图像对手势进行分割,并基于矩和链码进行手势质心与手势轨迹特征的计算,再利用动态时间规整算法进行手势轨迹特征识别,最后将识别结果传输给六足机器人进行人机交互实验,实现了动态手势对六足机器人的控制;实验结果证明:该方法识别准确率最高可达97%,且不易受光照影响,具有较强的鲁棒性,同时也满足了人机交互需求。

Research on dynamic gesture recognition based on Kinect

The traditional dynamic gesture recognition method based on vision in the field of human-computer interaction is not accurate enough and is easily affected by different intensity of light. Firstly, the depth image collected by Kinect sensor is used to segment the gesture, and the gesture centroid and gesture trajectory feature are calculated based on moment and chain code, and the gesture trajectory feature is recognized by dynamic time warping algorithm. Finally, the recognition results are transmitted to hexapod robot for human-computer interaction experiment, and the control of hexapod robot by dynamic gesture is realized. Experimental results show that the recognition accuracy of this method is up to 97% and is not affected by the light, and it has strong robustness and meets the requirements of human-computer interaction.