A machine learning based sentient multimedia framework to increase safety at work.

Area of Science:

• Computer Science
• Engineering
• Occupational Health

Background:

• Workplace safety has gained increasing importance, with Information and Communication Technology (ICT) playing a crucial role.
• The Internet of Things (IoT) and its architectures like Sentient Multimedia Systems (SMS) are vital for extracting predictive safety information.
• Machine Learning (ML) is frequently employed to analyze the vast data generated by these systems.

Purpose of the Study:

• To propose a novel framework integrating Sentient Multimedia Systems and Machine Learning for improved workplace safety.
• To demonstrate the framework's application in a specific scenario: fall detection.
• To present a practical, ML-based wearable device and a safety coordination platform.

Main Methods:

• Development of a framework combining SMS and ML for safety applications.
• Design, construction, and testing of an ML-based wearable device for fall detection.
• Implementation of a safety coordination platform for real-time monitoring and incident response.

Main Results:

• The proposed framework effectively utilizes SMS and ML for workplace safety.
• The wearable device accurately detects falls, demonstrating the system's efficacy.
• The safety platform successfully monitors the environment, triggers alarms, and provides guidance.

Conclusions:

• The integrated SMS and ML framework offers a robust solution for enhancing workplace safety.
• Wearable technology and intelligent platforms can significantly improve incident response, such as in fall detection.
• This approach has the potential to reduce workplace accidents and improve worker well-being.