An idea behind this AI-powered system
Working in a production hall usually involves wearing personal protective equipment (PPE), such as safety glasses, hard hats, reflective vests, or masks. Frequently more than one type of PPE at a time is required to keep staff safe.
The problem is that employees often unintentionally forget to wear them, which poses a serious danger not only to themselves but also to the companies they work for. However, as we were told by the company representatives, punishing workers for carelessness in this particular area simply doesn’t work. It is demotivating and thus counterproductive. Therefore, we needed to come up with a solution that would be effective in this matter, but ideally also wouldn’t engage too much human resources.
In a nutshell
- A system for automatic verification of whether workers wear masks
- The system can work 24/7, day and night shifts
- Data collection, analysis, and automated reporting
- A scalable solution that can be applied to other PPE types
More about the system
We have figured out that a direct reminder about wearing personal protective equipment could solve the problem. However, instead of delegating people to the task, which would be tedious for employees and costly for the company, we wanted the process to become automated and thus scalable. Thus, facing covid-related challenges, we have built an AI-based system that automatically verifies whether employees wear protective masks or not. To make it effective, we used artificial intelligence methods, such as deep neural networks, image processing, and object detection.
A camera mounted at the entrance of the production hall detects people with uncovered faces and displays a friendly reminder on the monitor: “Remember to wear a mask”. Also, the system says “Thank you for wearing a mask” to those who have it already on. The aim is to encourage workers to wear PPE and develop this habit in the long run. Most importantly, the solution uses positive motivation instead of any form of penalty.
The system can also collect statistical data, such as the number of employees not wearing masks. Additionally, it can specify the most sensitive time of the day. This allows managers to identify bottlenecks and monitor progress in time. The solution can work similarly with other types of PPE, being just another step in the company’s digital transformation.