摘要
基于传感器的人体行为识别是一个新兴研究领域,作为物联网的一项重要应用,在医疗监护、助老助残、智能办公/家居等方面有着广阔的应用前景。识别率是行为识别的一个重要衡量指标,而特征和分类算法又是影响识别率的两个重要因素。提取了基于多传感器行为识别架构的关联特征,并引入压缩感知和稀疏表示理论,提出一种多任务压缩感知行为识别方法。最后,在基准数据库上采用个体无关的留一验证方法进行了大量实验,结果表明所提出的融合关联性的多任务压缩感知行为识别方法能有效提升行为识别率,与对应的单任务行为识别方法相比,识别速度提高约56%。
Sensor-based human activity recognition is an emerging research field. It is an important application of Internet of Things （IoT）, and has very promising application prospects in health care/recovery, elder/invalid people assistant, smart home/office, etc. Accuracy is one of the most important evaluation standards of activity recognition, and appropriate features and classifiers are important accura- cy factors. We first extract a novel feature called correlation feature. By combining the theory of com- pressive sensing and sparse representation, we propose a multi-task compressive sensing method and use it as the classifier to resolve the problem of activity recognition. Finally, we conduct a large amount of experiments on a set of benchmarks with Leave-One-Subject-Out cross validation. Experimental results show that the extracted feature and the proposed method are effective in improving the accuracy of sen- sor-based activity recognition. Moreover, compared to the corresponding single task method, the pro- posed classifier can reduce the execution time by nearly 56 %.
出处
《计算机工程与科学》
CSCD
北大核心
2015年第6期1071-1078,共8页
Computer Engineering & Science
基金
国家自然科学基金资助项目(61173036
61272061)
湖南省科技计划资助项目(2014GK3009)
关键词
机器学习
物联网
体域网
行为识别
特征提取
machine learning
Internet of Things
body area network
activity recognition
feature extraction
