Human Behavior Recognition Model Based on Feature and Classifier Selection.

Ge Gao,Bangwen Zhou,Ying Chen,Zhixin Li,Chenhui Dong,Zhan Huan,Jiuzhen Liang

doi:10.3390/s21237791

Abstract

With the rapid development of the computer and sensor field, inertial sensor data have been widely used in human activity recognition. At present, most relevant studies divide human activities into basic actions and transitional actions, in which basic actions are classified by unified features, while transitional actions usually use context information to determine the category. For the existing single method that cannot well realize human activity recognition, this paper proposes a human activity classification and recognition model based on smartphone inertial sensor data. The model fully considers the feature differences of different properties of actions, uses a fixed sliding window to segment the human activity data of inertial sensors with different attributes and, finally, extracts the features and recognizes them on different classifiers. The experimental results show that dynamic and transitional actions could obtain the best recognition performance on support vector machines, while static actions could obtain better classification effects on ensemble classifiers; as for feature selection, the frequency-domain feature used in dynamic action had a high recognition rate, up to 99.35%. When time-domain features were used for static and transitional actions, higher recognition rates were obtained, 98.40% and 91.98%, respectively.

Highlights

The transitional actions between human continuous actions were balanced as samples and identified as basic actions
The basic actions were divided more carefully and different characteristics were extracted from the processed transition actions; the classification and recognition were compared on multiple learning classifiers
The experiments show that the oversampling of transitional actions was conducive to action recognition

Summary

Introduction

HAR has been widely used in medical care, social security, behavior analysis, traffic safety and other industries. In the field of health care, HAR has been successfully applied to behavior monitoring [1], observation [2] and classification [3] of the elderly, medical diagnosis of patients [4], rehabilitation and physical therapy [5]. In the field of social security, the detection of abnormal people [6] and the tracking of suspicious objects [7] are inseparable from HAR technology, which improves the precision attack ability against criminals [8]. In the behavioral analysis industry, HAR is used to monitor and identify family behavior [8]. In the field of traffic safety [9], the detection of driver fatigue is helpful to prevent traffic accidents caused by drowsiness during driving

Methods

Results

Conclusion