Bonifazi Gianluca, Corradini Enrico, Ursino Domenico, Virgili Luca, Anceschi Emiliano, De Donato Massimo Callisto
Department of Information Engineering, Polytechnic University of Marche, Ancona, Italy.
Gruppo Filippetti S.p.A., Ancona, Italy.
Multimed Tools Appl. 2022;81(1):141-169. doi: 10.1007/s11042-021-10984-z. Epub 2021 May 15.
In the last few decades, we have witnessed an increasing focus on safety in the workplace. ICT has always played a leading role in this context. One ICT sector that is increasingly important in ensuring safety at work is the Internet of Things and, in particular, the new architectures referring to it, such as SIoT, MIoT and Sentient Multimedia Systems. All these architectures handle huge amounts of data to extract predictive and prescriptive information. For this purpose, they often make use of Machine Learning. In this paper, we propose a framework that uses both Sentient Multimedia Systems and Machine Learning to support safety in the workplace. After the general presentation of the framework, we describe its specialization to a particular case, i.e., fall detection. As for this application scenario, we describe a Machine Learning based wearable device for fall detection that we designed, built and tested. Moreover, we illustrate a safety coordination platform for monitoring the work environment, activating alarms in case of falls, and sending appropriate advices to help workers involved in falls.
在过去几十年里，我们目睹了对工作场所安全的关注度日益提高。在这方面，信息通信技术（ICT）一直发挥着主导作用。在确保工作场所安全方面日益重要的一个ICT领域是物联网，尤其是与之相关的新架构，如社交物联网（SIoT）、移动物联网（MIoT）和感知多媒体系统。所有这些架构都处理大量数据以提取预测性和规范性信息。为此，它们经常利用机器学习。在本文中，我们提出了一个框架，该框架使用感知多媒体系统和机器学习来支持工作场所的安全。在对该框架进行总体介绍之后，我们将描述其针对特定案例（即跌倒检测）的专门化。对于此应用场景，我们描述了一款基于机器学习的用于跌倒检测的可穿戴设备，该设备是我们设计、制造并测试的。此外，我们还展示了一个安全协调平台，用于监控工作环境，在发生跌倒时触发警报，并向跌倒相关的工作人员发送适当的建议以提供帮助。
