Hand tremor identification method in non-static state of body

The invention discloses a hand tremor identification method in a non-static state of a body. The method is specifically implemented according to the following steps: 1, establishing a hand tremor mathematical model; 2, carrying out data calibration and noise elimination; 3, eliminating a time-varying direct current component and calculating a time-varying tremor frequency; and 4, carrying out a hand tremor identification method based on CNN-LSTM. The problem that in the prior art, the hand tremor action in the non-static state of the body cannot be accurately recognized is solved. 本发明公开了一种身体非静止状态的手部震颤识别方法，具体按照以下步骤实施：步骤1、建立手部震颤数学模型；步骤2、数据校准与噪音消除；步骤3、消除时变直流分量与计算时变震颤频率；步骤4、基于CNN-LSTM的手部震颤识别方法。本发明解决了现有技术中存在无法对身体非静止状态的手部震颤动作精确识别的问题。