期刊名称：International Journal of Artificial Intelligence & Applications (IJAIA)
印刷版ISSN：0976-2191
电子版ISSN：0975-900X
出版年度：2016
卷号：7
期号：6
页码：1
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：A method for human activity recognition using mobile phones is introduced. Using the accelerometer andgyroscope typically found in modern smartphones, a system that uses the proposed method is able torecognize low level activities, including athletic exercises, with high accuracy. A Hebbian learningpreprocessing stage is used to render accelerometer and gyroscope signals independent to the orientationof the smartphone inside the user’s pocket. After preprocessing, a selected set of features are obtained andused for classification by a k-nearest neighbor or a multilayer perceptron. The trained algorithm achievesan accuracy of 95.3 percent when using the multilayer perceptron and tested on unknown users who areasked to perform the exercises after placing the mobile device in their pocket without any constraints on theorientation. Comparison of performance with respect to other popular methods is provided.
关键词：accelerometer; gyroscope; human activity recognition; smartphone
