Many businesses are preparing to start working after the Coronavirus pandemic lockdown. In this situation, our top priority is bringing health and safety to the workplace. It is the foundation of everything we do. And our most important responsibility is to deliver reliable video vision products to provide safety at your workplace.

Beyond that, as AI specialists, we are committed to providing an additional way to help protect and serve the whole clients, and especially those who are most vulnerable.

During this pandemic, we want to reach out and update you on how we have approached the situation at Rhyton Solutions: The leading causes of COVID-19 spread are through respiratory droplets when an infected person coughs, sneezes or speaks. Also, the Occupational Safety and Health Administration (OSHA) advises encouraging respiratory etiquette, including covering coughs and sneezes.

Based on these risks, if employees have to be present at work, HSE regulation in most companies and industries advises using masks, keeping social distancing, and in case someone feels sick or has high fever, staying at home and not putting others in danger.

So we understand that many customers may be facing unexpected side effects of spreading this virus. To help, we provide an innovative self-detection solution for personal protective equipment, social distancing, fever detection, and face recognition under the mask using deep learning.

These tools can be used to detect the observance of safety principles and the necessary protective cover in working areas.

The input of the designed deep network is the images received from the cameras installed in the workplace. The used deep learning methods have a lightweight network with faster response speed and higher accuracy than before. The response time of the built-in tool is real-time, which results in efficient network performance in the workplace with a large number of cameras.

To optimize the accuracy and speed of the detection network, an innovative model has been designed in our intelligent EHS that optimizes the network dimension parameters, including the number of layers, network width, and input image resolution altogether.

Also, in the designed network, by using a combination of object features with different dimensions, we can improve detection performance in various scales.

A suitable dataset with various images has been created in our collection for detecting safety tools in the workplace as well as masks in outdoor environments to improve the performance.

Besides, The system can identify and track people in the workplace using the face recognition feature no matter wearing a mask or not.

Another important concern of OSHA is following social distancing protocols. We provide a new video vision social distancing detection tool designed to help monitor and enforce physical distancing protocols in workplaces.

The feature is to measure the social distancing between any two persons in the camera streams, using an innovative feature, which draws a virtual boundary box around each detected individual in the frame.

The next step is the measurement, in which the system scales the distance between each pair of boundary boxes, and depending on preset physical distancing minimums, indicates any non-compliant pair on the system display in real-time with a red frame and red line. Then it sends an alarm to the operator, ensuring they would not miss issues that require immediate attention.

To solve social distance tracking, we have two approaches;

First, The system enables us to give a perception of what is happening using event processing architecture with the help of object detection inside the intelligent EHS and the CEP(Complex Event Processing) unit of the platform to calculate people’s distance in real-time confidently.

And the CEP, on top of the big data technology, our innovative solution in Intelligent EHS, allows us to calculate the distance between objects in high volume. This method is a complex task due to the depth of the image.

The second approach is to use Deep Learning methods.

We are also working on training a Deep Learning Model to reach our satisfactory speed and accuracy for industrial usage that can be added easily to the system.

According to our researches, in most industrial companies, workers try to hide their illness to be more present to take advantage of overtime working.

It is very important for HSE managers that people suspected of having a fever do not show up at work at all.

On the other hand, the story of COVID-19 will probably remain with companies throughout this year.

Well-known manufacturers offer thermometer cameras that are provided in the network of CCTV cameras. Still, the problem is that the constant monitoring and human error itself is the cause of many mistakes that ultimately leads to the presence of a suspicious person in the workplace.

Using face recognition technology, along with thermal cameras, can assess the employees’ health in case of fever when they enter the workplace.

The intelligent EHS allows the EHS manager to change the threshold of the thermal camera to increase the sensitivity of the measurement for specific individuals, locations, or entry at different times according to the time and severity of the concern.

By using Intelligent EHS, factories and companies with a large number of workers can separate a certain number of employees by easily detecting suspicious individuals for further tests.

Furthermore, Intelligent EHS LPOB(Live People On Board) module provides the possibility to identify a person who is required to have a home quarantine but secretly intends to enter the company and informs the safety manager to follow the related HSE regulation.