This paper proposes a method of classification and representations for touch gestures based on meta-action and defines a set of stroke touch gestures for Human-Machine Interaction (HMI).A method of touch gesture training and recognition is presented based on RBF neural network.Experiment results show that the proposed method is reliable and efficient for the training and recognition of touch gestures, and can provide a more natural and intuitive human-machine interaction for devices with touch screen.%针对触摸显示屏的操作特点提出了一种基于元动作的触摸手势分类和表示方法,根据人机交互要求定义了一套笔画触摸手势,提出了基于RBF神经网络的笔画触摸手势训练和识别方法.测试结果表明,所提出的方法能够快速、准确地对触摸手势进行训练和识别,可以为带触摸屏的设备提供一个更加自然、直观的人机交互手段.
展开▼