Daily activity identifying method with exercising and physiology sensing data fused

Abstract

The invention discloses a daily activity identifying method with exercising and physiology sensing data fused. According to the method, an smart phone and a wearable physiology sensing device are used for collecting exercising sensing data and physiology sensing data; then time domain and frequency domain statistical features and nonlinear features are extracted respectively, and feature selection is carried out through a sequence floating forward selection method; through a support vector machine and a Gaussian mixture model, activity identifying submodels of the exercising sensing data and the physiology sensing data are trained respectively; and finally the submodels are subjected to weighting integration through a fractional order fusion method, and a final daily activity classification model is obtained. The exercising sensing data and the physiology sensing data are fused, the accuracy of daily activity identifying can be improved, and the daily activity identifying method has wide application prospect in the fields of intelligent families, medial health care, old people assisting and the like.

Description

technical field [0001] The invention relates to the fields of pattern recognition and pervasive computing, in particular to a daily activity recognition method that integrates motion and physiological sensor data. Background technique [0002] With the development of smart phones and wearable sensors, the acquisition of motion data such as position, acceleration, angular velocity, and orientation, as well as physiological sensor data such as ECG, respiration, and body temperature has become increasingly convenient. How to use these data and signals for daily activities Identification has become the focus of attention in the industry, and the corresponding technology has broad application prospects in the fields of smart home, medical care, and assistance for the elderly. [0003] Existing activity recognition methods generally only recognize sports-related activities based on acceleration sensing data, such as standing, sitting, walking, running, cycling, etc. [0004] The ...

AI Technical Summary

Problems solved by technology

Aiming at the problem that it is difficult to effectively integrate motion data and physiological sensing data at the signal level and feature level, the method of the present invention performs feature extraction on the motion data and physiological sensing data respectively, and then uses the sequential forward floating search strategy for the extracted features as features Then, the obtained eigenvectors are respectively established through the Gaussian mixture model in the generative model and the support vector machine in the discriminative model to establish a classification sub-model. Finally, the decision-making of these classification sub-models is fused by the method of fractional fusion and the final output decision making

Images