In recent months, the regular cadence of life has been greatly disrupted. As a result, many companies have had to close their doors to employees and customers alike and have instead relied on remote working to maintain business operations. As retailers and other businesses reopen, many organizations are using a combination of testing, fever screening and contact tracing to create safer environments for employees, customers and suppliers.
What is the situation?
Organizations have begun to reopen to foot traffic, and many essential employees are returning to the workplace. As this happens, companies will need to ensure those entering the workplace are doing so safely—screening to ensure those who may be sick are not potentially exposing others, ensuring that those who do enter are adhering to social distancing recommendations, protecting those performing the screenings from potentially becoming infected themselves and identifying those mostly asymptomatic carriers who may be unknowingly spreading the virus to others.
To help provide enhanced public health screening to companies around the world, WWT and NVIDIA have partnered with IntelliSite, a company on the leading edge of delivering AI and IoT services. WWT, NVIDIA and IntelliSite together are deploying human-based monitoring (hBM) systems to detect elevated human body temperatures.
The partnership is highly strategic in nature. Combining IntelliSite’s host of hardware and software solutions and WWT’s expertise in integration and deployment, along with NVIDIA’s GPU offerings, the three companies are set to deliver two innovative solutions to help mitigate against the risks in re-opening buildings: facilities and work locations.
What are the solutions?
IntelliSite developed its hBM system from the software, hardware and connectivity technologies created since their inception in 2001. The hBM system combines four key technologies to provide safe and efficient temperature readings: thermal imaging, artificial intelligence (AI), managed services and personalized portals. The system provides accurate temperature readings using a Mobotix camera with two lenses, a color RGB lens and a thermographic lens. Together, the two lenses provide screening agents with a complete picture of the scene.
Secondly, the system uses AI to process data on-premise without it having to be transmitted to the cloud. It uses Deep Vision AI at the edge to analyze the information in real-time, provide the average, max and min temperatures of individuals in the field of vision and trigger an alarm if a temperature is detected above the threshold. In addition, the system will store the image of those individuals whose temperatures register above targeted ranges.
Figure 1. hBM high-level design and architecture
Thirdly, the system provides managed services, which ensures that the remote agents will be able to provide system updates and related support to the hBM system as needed. Finally, the system’s personalized portal for managerial personnel allows administrators access to metrics regarding the system’s event activity (e.g., total scans in given period, number of alarms per hour).
Human-Based Monitoring - Advanced (hBM-X)
These features are integrated together to comprise the crux of the hBM solution. Within the family of hBM solutions sits the hBM-X solution offering. The hBM-X solution provides customers with the ability to visually detect both objects and persons with temperature differences of as little as 0.1 degrees Celsius. It comes with the other features central to the hBM central product offering, features which include:
- alarms for the notification of temperatures either exceeding or falling below defined temperature ranges;
- event triggers (e.g., alarm, network message, activation of a switching output);
- screening via special TR windows or the complete sensor image; and
- a temperature range from -40 to +550 degrees Celsius.
To perform most of the use cases that customers are demanding, the hBM-X provides an effective solution. Customers will also receive a mobile cart to mount the components, a touchscreen display, cloud-based services and remote device management.
The hBM-X uses NVIDIA’s Quadro GPUs to provide the extra processing power required to deliver real-time analytics and low latency visualization. Together, these features offer a scalable, safe and efficient method by which companies can appropriately introduce people back into their physical operating space.
Figure 2. An example of WWT and IntelliSite’s hBM-X mobile cart solution components
Human-Based Monitoring - Enterprise (hBM-E)
For companies looking to address larger enterprise-wide use cases, WWT offers the hBM-E solution set. Like the hBM-X, the hBM-E provides fixed Mobotix camera installation across facilities processing AI and machine learning (ML) models based on the use cases and outcomes the customer desires. Its architecture uses a combination of edge processing, data center connectivity, enterprise-wide visualization and data centralization to deliver a dynamic solution offering for companies.
The hBM-E system also provides companies the ability to process their data and perform their analytics in a private cloud instance, keeping it out of third-party data centers. This can be particularly attractive for customers (e.g., hospitals, government facilities) who need greater protection around their data.
Figure 3. hBM-E offers a dynamic solution for enterprise-wide use cases across various verticals
Although this solution addresses a very particular use case, it can also be used to solve adjacent use cases. This solution is essentially an AI-enabled IoT device at the edge fitted to analyze thermal images and color images for elevated body temperature. However, it can also be applied to other analytics use cases – more than 100 have been identified - where sensors and cameras are leveraged.
For example, using Deep Vision’s AI software and appropriate data sources, the system may infer the relative age of an individual as well as their identity. Its ability to perform pattern recognition applies not just to people but also to many other objects such as buildings, vehicles and signage.
In addition to healthcare, other industries are also using object recognition, as well as thermal recognition, to provide actionable outcomes. Retailers can use that information to provide customers with specific recommendations and material to improve their shopping experience.
In other domains, integration with role-based access control system can provide notice of illegal entry and enforce compliance. This solution will therefore have a great opportunity to address other pressing problems outside of the immediate situation to help individuals, companies and governments provide better service for their customers and employees.
With hBM, organizations now have a scalable solution to create safer spaces, more rapidly and safely screen individuals at entry points, and bring their employees and customers together again in public spaces.
A challenge that most companies have experienced with manual fever screening protocols is that they often create a huge bottleneck of individuals waiting to have their temperatures checked. This is not only inefficient but it poses greater health risks for people not standing six feet apart while doing so. By screening those people who appear most at risk, coupled with random screening protocols, companies will do their part in keeping the public safe going forward.
This strategic partnership between WWT, NVIDIA and IntelliSite delivers a proven solution for reintroducing people back into the workplace and combines the strengths of each company to provide a much needed and timely service for safer spaces.