基于RBF神经网络的人类活动识别

Human activity recognition based on radial basis function neural network

为了实现对智能家居中人类日常生活活动(ADLS)的识别,将使用径向基函数RBF神经网络来进行人类活动的识别。并使用志愿者在智能家居试验台执行活动搜集到的数据对算法的准确率进行评估。实验结果表明,选择合适的特征量和参数,相比于隐含马尔科夫模型径向基函数神经网络人类活动的识别方面显示了较高的准确率。

In order to recognize the activities of daily living ADLS in smart home.In this paper,RBF neural network is applied to recognize the human activities.To evaluate the accuracy of the recognition algorithms,the results using real data collected from participants performing activities were assessed.With proper feature selections,the results of radial basis function neural network show the significant ability to recognize human activities in smart home.