Recognizing Seatbelt-Fastening Behavior with Wearable Technology and Machine Learning
Recognizing Seatbelt-Fastening Behavior with Wearable Technology and Machine Learning
复制标题
DOI：
10.1145/3411763.3451705
复制
发表时间：
2021-05
期刊：
影响因子：
--
通讯作者：
J. M. Leland;Ellen Stanfill;Josh Cherian;T. Hammond
中科院分区：
文献类型：
--
作者：
J. M. Leland;Ellen Stanfill;Josh Cherian;T. Hammond
每年有近135万人死于汽车事故，而在这些事故中，近一半的人在事故发生时没有系安全带。尽管现代车辆中嵌入了大量的安全传感器和警告指示器，但仍然缺乏安全预防措施。这就提出了一个明确的需要，以更有效的方法，鼓励一贯使用安全带。为此，这项工作利用可穿戴技术和活动识别技术来检测人们何时系上安全带。为了开发这样一个系统，我们收集了26个不同用户的智能手表数据。从这些数据中，我们确定了激发新功能开发的趋势。利用这些特征，我们训练模型来实时识别系安全带的动作。这一模型是未来工作的基础，在未来的工作中，系统可以提供个性化和有效的干预措施，以确保安全带的使用。
Nearly 1.35 million people are killed in automobile accidents every year, and nearly half of all individuals involved in these accidents were not wearing their seatbelt at the time of the crash. This lack of safety precaution occurs in spite of the numerous safety sensors and warning indicators embedded within modern vehicles. This presents a clear need for more effective methods of encouraging consistent seatbelt use. To that end, this work leverages wearable technology and activity recognition techniques to detect when individuals have buckled their seatbelt. To develop such a system, we collected smartwatch data from 26 different users. From this data, we identified trends which inspired the development of novel features. Using these features, we trained models to identify the motion of fastening a seatbelt in real-time. This model serves as the basis for future work in which systems can provide personalized and effective interventions to ensure seatbelt use.