基于加速度传感器和神经网络的人体活动行为识别

Human activity behavior recognition based on acceleration sensor and neural network

人体活动行为识别在医疗、安全、娱乐等方面有着广泛的应用,为了高效、准确地获取人体活动的行为信息,提出一种基于加速度传感器和神经网络的个人活动行为识别方法。该方法通过在个人手上佩戴加速度传感器,实时采集个人活动的行为数据;再通过BP神经网络分析相关行为数据并建立个人活动行为模型,分类识别个人的行走、坐着、躺卧、站立和突然跌倒等活动行为特征。实验结果表明,该方法能够有效检测到个人活动的行为特征参数,并可准确识别出人体活动的五种典型行为。

Human activity behavior recognition has a wide range of applications in medical treatment,safety,entertainment and so on. In order to efficiently and accurately obtain human activity behavior information,a method of personal activity behavior recognition based on acceleration sensor and neural network is proposed in this paper. The method can realize realtime acquisition of personal activity behavior data by acceleration sensor acquisition nodes worn on the hands of individuals. The relevant behavior data is analyzed and the personal activity behavior model is established by means of BP neural network,so as to identify the characteristics of five kinds of individual activity behaviors:walking,sitting,lying down,standing and suddenly falling according to classification criteria. The experimental results show that the proposed method can effectively detect the behavioral characteristic parameters of individual activities and accurately identify five typical behaviors of human activities.